[go: up one dir, main page]

CN106355429A - Image material recommendation method and device - Google Patents

Image material recommendation method and device Download PDF

Info

Publication number
CN106355429A
CN106355429A CN201610676576.9A CN201610676576A CN106355429A CN 106355429 A CN106355429 A CN 106355429A CN 201610676576 A CN201610676576 A CN 201610676576A CN 106355429 A CN106355429 A CN 106355429A
Authority
CN
China
Prior art keywords
image material
recommended
keywords
text information
advertisement image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201610676576.9A
Other languages
Chinese (zh)
Inventor
吴宏刚
闫铭
陆子龙
张卫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Xiaomi Mobile Software Co Ltd
Original Assignee
Beijing Xiaomi Mobile Software Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Xiaomi Mobile Software Co Ltd filed Critical Beijing Xiaomi Mobile Software Co Ltd
Priority to CN201610676576.9A priority Critical patent/CN106355429A/en
Publication of CN106355429A publication Critical patent/CN106355429A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0251Targeted advertisements
    • G06Q30/0255Targeted advertisements based on user history
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0277Online advertisement

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention provides a image material recommendation method and device, comprising: acquiring a text information entered by the users; generating a keyword associated with the text information; acquiring the recommended advertisement image material according to the keyword; and outputs the recommended advertisement image material. The image material recommendation method and device can recommend the advertisement material material to the user according to the text information input by the user, and provides the user with more reference material, thus improving the efficiency and effect of advertisement delivery.

Description

图像素材的推荐方法及装置Recommended method and device for image material

技术领域technical field

本公开涉及计算机技术领域,特别涉及一种图像素材的推荐方法及装置。The present disclosure relates to the field of computer technology, in particular to a method and device for recommending image materials.

背景技术Background technique

随着网络技术的不断发展,互联网已经成为人们生活和工作中必不可少的一种信息传播媒介。因此,很多广告主会将自己产品的广告投放到互联网中。产品的广告可以包括很多素材,例如文本素材、图像素材以及声音素材等。其中,广告的图像素材对广告的点击率有较大的影响,有吸引力的图像能够引导更多的用户点击广告,增加广告的点击率。With the continuous development of network technology, the Internet has become an indispensable information medium in people's life and work. Therefore, many advertisers will place advertisements of their products on the Internet. The advertisement of a product may include many materials, such as text material, image material, and sound material. Among them, the image material of the advertisement has a greater influence on the click-through rate of the advertisement, and an attractive image can guide more users to click on the advertisement and increase the click-through rate of the advertisement.

目前,一些广告主在投放某些产品的广告时,一般只能通过调研以及凭借自己的经验,为所要投放的广告选择图像素材,并通过广告平台将这些产品的广告投放出去。因此,在图像素材的选择上需要花费大量的时间,使得广告投放的效率低下,效果较差。At present, when some advertisers place advertisements for certain products, they generally can only select image materials for the advertisements to be placed through research and their own experience, and place the advertisements for these products through the advertising platform. Therefore, it takes a lot of time to select the image material, so that the efficiency of advertisement delivery is low and the effect is poor.

发明内容Contents of the invention

为了解决上述技术问题,本公开提供一种图像素材的推荐方法及装置。In order to solve the above technical problems, the present disclosure provides a method and device for recommending image materials.

根据本公开实施例的第一方面,提供一种图像素材的推荐方法,包括:According to the first aspect of the embodiments of the present disclosure, a method for recommending image materials is provided, including:

获取用户输入的文本信息;Obtain the text information entered by the user;

生成与所述文本信息关联的关键词;generating keywords associated with the text information;

根据所述关键词获取推荐的广告图像素材;Acquiring recommended advertisement image materials according to the keywords;

输出所述推荐的广告图像素材。Outputting the recommended advertisement image material.

可选的,所述生成与所述文本信息关联的关键词,包括:Optionally, said generating keywords associated with said text information includes:

从预存的相关词汇集合中获取与所述文本信息关联的相关词汇集合;Acquiring a related vocabulary set associated with the text information from a pre-stored related vocabulary set;

将与所述文本信息关联的相关词汇集合中包括的词语作为所述关键词。The words included in the related vocabulary set associated with the text information are used as the keywords.

可选的,所述生成与所述文本信息关联的关键词,包括:Optionally, said generating keywords associated with said text information includes:

获取预先训练的主题模型;Get a pre-trained topic model;

将所述文本信息输入所述主题模型,以生成主题词语作为所述关键词。The text information is input into the topic model to generate topic words as the keywords.

可选的,所述根据所述关键词获取推荐的广告图像素材,包括:Optionally, the acquiring recommended advertising image material according to the keywords includes:

采用网页爬虫技术从互联网络中爬取与所述关键词关联的广告图像素材作为推荐的广告图像素材。Advertisement image materials associated with the keywords are crawled from the Internet by web crawler technology as recommended advertisement image materials.

可选的,所述根据所述关键词获取推荐的广告图像素材,包括:Optionally, the acquiring recommended advertising image material according to the keywords includes:

将预先建立的广告图像素材库中与所述关键词匹配的广告图像素材作为推荐的广告图像素材。Advertisement image materials matching the keywords in the pre-established advertisement image material library are used as recommended advertisement image materials.

可选的,所述方法还包括:Optionally, the method also includes:

获取互联网络中的历史搜索记录数据;Obtain historical search record data in the Internet;

基于所述历史搜索记录数据生成多个相关词汇集合;generating a plurality of related vocabulary sets based on the historical search record data;

将生成的多个相关词汇集合进行存储。Store the generated sets of related vocabulary.

可选的,所述基于所述历史搜索记录数据生成多个相关词汇集合,包括:Optionally, the generating a plurality of related vocabulary sets based on the historical search record data includes:

采用SimRank算法对所述历史搜索记录数据进行分类处理,以生成多个相关数据集合。The SimRank algorithm is used to classify the historical search record data to generate a plurality of related data sets.

可选的,其特征在于,所述方法还包括:Optionally, it is characterized in that the method also includes:

获取参考广告图像素材以及所述参考广告图像素材对应的关键词;Obtaining a reference advertisement image material and keywords corresponding to the reference advertisement image material;

将所述参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。The reference advertisement image material is stored in association with corresponding keywords to establish an advertisement image material library.

可选的,所述输出所述推荐的广告图像素材,包括:Optionally, the outputting the recommended advertising image material includes:

分别将所述推荐的广告图像素材输入到预先训练的广告图像素材质量评估模型,以获取所述推荐的广告图像素材对应的质量级别;respectively inputting the recommended advertising image material into a pre-trained advertising image material quality evaluation model to obtain the quality level corresponding to the recommended advertising image material;

将所述推荐的广告图像素材按照质量级别从高到低的顺序进行排序;sorting the recommended advertising image materials in order of quality level from high to low;

按照所述排序输出所述推荐的广告图像素材。Outputting the recommended advertisement image material according to the ranking.

根据本公开实施例的第二方面,提供一种图像素材的推荐装置,包括:According to the second aspect of the embodiments of the present disclosure, an apparatus for recommending image materials is provided, including:

第一获取模块,被配置为获取用户输入的文本信息;The first obtaining module is configured to obtain the text information input by the user;

生成模块,被配置为生成与所述文本信息关联的关键词;a generation module configured to generate keywords associated with the text information;

第二获取模块,被配置为根据所述关键词获取推荐的广告图像素材;The second acquiring module is configured to acquire the recommended advertisement image material according to the keywords;

输出模块,被配置为输出所述推荐的广告图像素材。An output module configured to output the recommended advertising image material.

可选的,所述生成模块包括:Optionally, the generating module includes:

第一获取子模块,被配置为从预存的相关词汇集合中获取与所述文本信息关联的相关词汇集合;The first acquiring submodule is configured to acquire a related vocabulary set associated with the text information from a pre-stored related vocabulary set;

确定子模块,被配置为将与所述文本信息关联的相关词汇集合中包括的词语作为所述关键词。The determining submodule is configured to use the words included in the related vocabulary set associated with the text information as the keywords.

可选的,所述生成模块包括:Optionally, the generating module includes:

第二获取子模块,被配置为获取预先训练的主题模型;The second obtaining submodule is configured to obtain a pre-trained topic model;

生成子模块,被配置为将所述文本信息输入所述主题模型,以生成主题词语作为所述关键词。A generating submodule configured to input the text information into the topic model to generate topic words as the keywords.

可选的,所述第二获取模块包括:Optionally, the second acquisition module includes:

爬取子模块,被配置为采用网页爬虫技术从互联网络中爬取与所述关键词关联的广告图像素材作为推荐的广告图像素材。The crawling sub-module is configured to crawl the advertising image material associated with the keyword from the Internet as a recommended advertising image material by using web crawler technology.

可选的,所述第二获取模块包括:Optionally, the second acquisition module includes:

推荐子模块,被配置为将预先建立的广告图像素材库中与所述关键词匹配的广告图像素材作为推荐的广告图像素材。The recommending sub-module is configured to use the advertising image material matching the keyword in the pre-established advertising image material library as the recommended advertising image material.

可选的,所述装置还包括:Optionally, the device also includes:

第三获取模块,被配置为获取互联网络中的历史搜索记录数据;The third obtaining module is configured to obtain historical search record data in the Internet;

集合生成模块,被配置为基于所述历史搜索记录数据生成多个相关词汇集合;A collection generating module configured to generate a plurality of related vocabulary collections based on the historical search record data;

存储模块,被配置为将生成的多个相关词汇集合进行存储。The storage module is configured to store the generated sets of related vocabulary.

可选的,所述集合生成模块包括:Optionally, the set generation module includes:

处理子模块,被配置为采用SimRank算法对所述历史搜索记录数据进行分类处理,以生成多个相关数据集合。The processing sub-module is configured to use the SimRank algorithm to classify and process the historical search record data to generate a plurality of related data sets.

可选的,所述装置还包括:Optionally, the device also includes:

第四获取模块,被配置为获取参考广告图像素材以及所述参考广告图像素材对应的关键词;The fourth obtaining module is configured to obtain a reference advertisement image material and keywords corresponding to the reference advertisement image material;

建立模块,被配置为将所述参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。The establishment module is configured to store the reference advertisement image material in association with corresponding keywords, so as to establish an advertisement image material library.

可选的,所述输出模块包括:Optionally, the output module includes:

输入子模块,被配置为分别将所述推荐的广告图像素材输入到预先训练的广告图像素材质量评估模型,以获取所述推荐的广告图像素材对应的质量级别;The input sub-module is configured to respectively input the recommended advertising image material into the pre-trained advertising image material quality evaluation model, so as to obtain the quality level corresponding to the recommended advertising image material;

排序子模块,被配置为将所述推荐的广告图像素材按照质量级别从高到低的顺序进行排序;The sorting submodule is configured to sort the recommended advertising image materials in order of quality level from high to low;

输出子模块,被配置为按照所述排序输出所述推荐的广告图像素材。The output sub-module is configured to output the recommended advertising image material according to the ranking.

根据本公开实施例的第三方面,提供一种图像素材的推荐装置,包括:According to a third aspect of the embodiments of the present disclosure, an apparatus for recommending image materials is provided, including:

处理器;processor;

用于存储处理器可执行指令的存储器;memory for storing processor-executable instructions;

其中,所述处理器被配置为:Wherein, the processor is configured as:

获取用户输入的文本信息;Obtain the text information entered by the user;

生成与所述文本信息关联的关键词;generating keywords associated with the text information;

根据所述关键词获取推荐的广告图像素材;Acquiring recommended advertisement image materials according to the keywords;

输出所述推荐的广告图像素材。Outputting the recommended advertisement image material.

本公开的实施例提供的技术方案可以包括以下有益效果:The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects:

本公开的上述实施例提供的图像素材的推荐方法,通过获取用户输入的文本信息,生成与该文本信息关联的关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,提高了广告投放的效率及效果。The image material recommendation method provided by the above-mentioned embodiments of the present disclosure obtains the text information input by the user, generates a keyword associated with the text information, obtains the recommended advertisement image material according to the keyword, and outputs the recommended advertisement image material . Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, more reference materials can be provided for the user, and the efficiency and effect of advertisement placement can be improved.

本公开的上述实施例提供的图像素材的推荐方法,可以获取用户输入的文本信息,从预存的相关词汇集合中获取与上述文本信息关联的相关词汇集合,将与该文本信息关联的相关词汇集合中包括的词语作为关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高广告投放的效率及效果。The image material recommendation method provided by the above-mentioned embodiments of the present disclosure can obtain the text information input by the user, obtain the related vocabulary set associated with the above-mentioned text information from the pre-stored related vocabulary set, and store the related vocabulary set associated with the text information The words included in are used as keywords, the recommended advertising image material is obtained according to the keyword, and the recommended advertising image material is output. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

本公开的上述实施例提供的图像素材的推荐方法,可以获取用户输入的文本信息,再获取预先训练的主题模型,将上述文本信息输入该主题模型,以生成主题词语作为关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高广告投放的效率及效果。The image material recommendation method provided by the above-mentioned embodiments of the present disclosure can obtain the text information input by the user, and then obtain the pre-trained topic model, input the above-mentioned text information into the topic model, and generate topic words as keywords. The word obtains the recommended advertising image material, and outputs the recommended advertising image material. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

本公开的上述实施例提供的图像素材的推荐方法,通过获取用户输入的文本信息,生成与该文本信息关联的关键词,采用网页爬虫技术从互联网络中爬取与该关键词关联的广告图像素材作为推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,从更丰富的广告图像素材中获取推荐广告图像素材,并向用户推荐广告图像素材,进一步为用户提供更多的参考素材,并提高了广告投放的效率及效果。The image material recommendation method provided by the above-mentioned embodiments of the present disclosure obtains the text information input by the user, generates keywords associated with the text information, and uses web crawler technology to crawl the advertisement images associated with the keywords from the Internet. The material is used as the recommended advertising image material, and the recommended advertising image material is output. Therefore, according to the text information input by the user, the recommended advertising image material can be obtained from more abundant advertising image materials, and the advertising image material can be recommended to the user, further providing more reference materials for the user, and improving the efficiency and effect of advertising placement .

本公开的上述实施例提供的图像素材的推荐方法,通过获取用户输入的文本信息,生成与该文本信息关联的关键词,将预先建立的广告图像素材库中与上述关键词匹配的广告图像素材作为推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,获取较高质量的推荐广告图像素材,并向用户推荐广告图像素材,为用户提供更多的参考素材,并进一步提高了广告投放的效率及效果。The method for recommending image materials provided by the above-mentioned embodiments of the present disclosure obtains the text information input by the user, generates keywords associated with the text information, and selects the advertisement image materials in the pre-established advertisement image material database that match the above keywords As a recommended advertising image material, and output the recommended advertising image material. Therefore, according to the text information input by the user, higher-quality recommended advertisement image materials can be obtained, and the advertisement image materials can be recommended to the user, providing more reference materials for the user, and further improving the efficiency and effect of advertisement placement.

本公开的上述实施例提供的图像素材的推荐方法,通过基于获取到的历史搜索记录数据生成多个相关词汇集合,再获取用户输入的文本信息,从预存的相关词汇集合中获取与文本信息关联的相关词汇集合,将与文本信息关联的相关词汇集合中包括的词语作为关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高广告投放的效率及效果。The method for recommending image materials provided by the above-mentioned embodiments of the present disclosure generates a plurality of sets of related words based on the acquired historical search record data, and then acquires the text information input by the user, and obtains the related words associated with the text information from the pre-stored set of related words. A set of related words, using words included in the set of related words associated with the text information as keywords, obtaining recommended advertising image materials according to the keywords, and outputting recommended advertising image materials. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

本公开的上述实施例提供的图像素材的推荐方法,通过获取参考广告图像素材以及参考广告图像素材对应的关键词,将参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。再获取用户输入的文本信息,生成与该文本信息关联的关键词,并将预先建立的上述广告图像素材库中与上述关键词匹配的广告图像素材作为推荐的广告图像素材,输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高了广告投放的效率及效果。The image material recommendation method provided by the above-mentioned embodiments of the present disclosure obtains the reference advertisement image material and the keywords corresponding to the reference advertisement image material, and associates and stores the reference advertisement image material and the corresponding keyword to create an advertisement image material library. Then obtain the text information input by the user, generate keywords associated with the text information, and use the advertisement image material matching the above keyword in the pre-established advertisement image material library as the recommended advertisement image material, and output the recommended advertisement image material. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

本公开的上述实施例提供的图像素材的推荐方法,按照质量级别从高到低的顺序对获取到的推荐的广告图像素材进行排序,并按照上述排序输出推荐的广告图像素材。从而可以使用户优先看到质量较高的广告图像素材,并为用户提供更多的参考素材,进一步提高了广告投放的效率及效果。The method for recommending image materials provided by the above-mentioned embodiments of the present disclosure sorts the obtained recommended advertising image materials in descending order of quality levels, and outputs the recommended advertising image materials according to the above-mentioned sorting. Therefore, the user can preferentially see the high-quality advertisement image material, and provide the user with more reference materials, further improving the efficiency and effect of advertisement placement.

应当理解的是,以上的一般描述和后文的细节描述仅是示例性和解释性的,并不能限制本公开。It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure.

附图说明Description of drawings

此处的附图被并入说明书中并构成本说明书的一部分,示出了符合本公开的实施例,并与说明书一起用于解释本公开的原理。The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description serve to explain the principles of the disclosure.

图1是本公开根据一示例性实施例示出的一种图像素材的推荐方法的流程图;Fig. 1 is a flowchart of a method for recommending image materials according to an exemplary embodiment of the present disclosure;

图2是本公开根据一示例性实施例示出的另一种图像素材的推荐方法的流程图;Fig. 2 is a flowchart of another method for recommending image materials according to an exemplary embodiment of the present disclosure;

图3是本公开根据一示例性实施例示出的另一种图像素材的推荐方法的流程图;Fig. 3 is a flowchart of another method for recommending image materials according to an exemplary embodiment of the present disclosure;

图4是本公开根据一示例性实施例示出的一种图像素材的推荐装置的框图;Fig. 4 is a block diagram of an apparatus for recommending image materials according to an exemplary embodiment of the present disclosure;

图5是本公开根据一示例性实施例示出的另一种图像素材的推荐装置的框图;Fig. 5 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure;

图6是本公开根据一示例性实施例示出的另一种图像素材的推荐装置的框图;Fig. 6 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure;

图7是本公开根据一示例性实施例示出的另一种图像素材的推荐装置的框图;Fig. 7 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure;

图8是本公开根据一示例性实施例示出的另一种图像素材的推荐装置的框图;Fig. 8 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure;

图9是本公开根据一示例性实施例示出的另一种图像素材的推荐装置的框图;Fig. 9 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure;

图10是本公开根据一示例性实施例示出的另一种图像素材的推荐装置的框图;Fig. 10 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure;

图11是本公开根据一示例性实施例示出的另一种图像素材的推荐装置的框图;Fig. 11 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure;

图12是本公开根据一示例性实施例示出的另一种图像素材的推荐装置的框图;Fig. 12 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure;

图13是本公开根据一示例性实施例示出的一种图像素材的推荐装置的一结构示意图。Fig. 13 is a schematic structural diagram of an apparatus for recommending image materials according to an exemplary embodiment of the present disclosure.

具体实施方式detailed description

这里将详细地对示例性实施例进行说明,其示例表示在附图中。下面的描述涉及附图时,除非另有表示,不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本公开相一致的所有实施方式。相反,它们仅是与如所附权利要求书中所详述的、本公开的一些方面相一致的装置和方法的例子。Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatuses and methods consistent with aspects of the present disclosure as recited in the appended claims.

在本公开使用的术语是仅仅出于描述特定实施例的目的,而非旨在限制本公开。在本公开和所附权利要求书中所使用的单数形式的“一种”、“所述”和“该”也旨在包括多数形式,除非上下文清楚地表示其他含义。还应当理解,本文中使用的术语“和/或”是指并包含一个或多个相关联的列出项目的任何或所有可能组合。The terminology used in the present disclosure is for the purpose of describing particular embodiments only, and is not intended to limit the present disclosure. As used in this disclosure and the appended claims, the singular forms "a", "the", and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It should also be understood that the term "and/or" as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

应当理解,尽管在本公开可能采用术语第一、第二、第三等来描述各种信息,但这些信息不应限于这些术语。这些术语仅用来将同一类型的信息彼此区分开。例如,在不脱离本公开范围的情况下,第一信息也可以被称为第二信息,类似地,第二信息也可以被称为第一信息。取决于语境,如在此所使用的词语“如果”可以被解释成为“在……时”或“当……时”或“响应于确定”。术语“纵向”、“横向”、“上”、“下”、“前”、“后”、“左”、“右”、“竖直”、“水平”、“顶”、“底”“内”、“外”等指示的方位或位置关系为基于附图所示的方位或位置关系,仅是为了便于描述本公开和简化描述,而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作,因此不能理解为对本公开的限制。It should be understood that although the terms first, second, third, etc. may be used in the present disclosure to describe various information, the information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of the present disclosure, first information may also be called second information, and similarly, second information may also be called first information. Depending on the context, the word "if" as used herein may be interpreted as "at" or "when" or "in response to a determination." The terms "portrait", "landscape", "top", "bottom", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom"" The orientation or positional relationship indicated by "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present disclosure and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, are constructed and operate in a particular orientation and therefore are not to be construed as limitations of the present disclosure.

需要说明的是,除非另有规定和限定,术语“安装”、“相连”、“连接”应做广义理解,例如,可以是机械连接或电连接,也可以是两个元件内部的连通,可以是直接相连,也可以通过中间媒介间接相连,对于本领域的普通技术人员而言,可以根据具体情况理解所述术语的具体含义。It should be noted that, unless otherwise specified and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense, for example, it can be a mechanical connection or an electrical connection, or it can be the internal communication of two components, and it can be It is directly connected, or indirectly connected through an intermediary, and those skilled in the art can understand the specific meanings of the terms according to specific situations.

如图1所示,图1是根据一示例性实施例示出的一种图像素材的推荐方法的流程图,该方法可以应用于终端设备或者服务器中。本领域技术人员可以理解,该终端设备可以包括但不限于诸如智能手机的移动终端设备、平板电脑、膝上型便携计算机以及台式电脑等等。该方法包括以下步骤:As shown in FIG. 1 , FIG. 1 is a flow chart of a method for recommending image materials according to an exemplary embodiment, and the method can be applied to a terminal device or a server. Those skilled in the art can understand that the terminal device may include, but is not limited to, a mobile terminal device such as a smart phone, a tablet computer, a laptop computer, a desktop computer, and the like. The method includes the following steps:

在步骤101中,获取用户输入的文本信息。In step 101, text information input by a user is obtained.

一般来说,产品的广告可以包括很多素材,例如文本素材、图像素材以及声音素材等。其中,图像素材即为图像形式的素材。在本实施例中,当用户想要获取为某产品推荐的广告图像素材时,需要首先输入和该产品相关的文本信息。例如,输入产品的名称,产品的属性,产品的特征,产品的类别以及产品的简介和描述等等,以便获取为该产品推荐的广告图像素材。因此,上述文本信息为与待投放广告对应的产品相关的一些文本信息。可以理解,本公开对文本信息的具体形式和内容方面不限定。Generally speaking, an advertisement of a product may include many materials, such as text material, image material, and sound material. Wherein, the image material is a material in the form of an image. In this embodiment, when the user wants to obtain the recommended advertising image material for a certain product, he needs to first input text information related to the product. For example, the name of the product, the attribute of the product, the feature of the product, the category of the product, the introduction and description of the product, etc. are input, so as to obtain the recommended advertising image material for the product. Therefore, the above text information is some text information related to the product corresponding to the advertisement to be placed. It can be understood that the present disclosure does not limit the specific form and content of the text information.

在步骤102中,生成与该文本信息关联的关键词。In step 102, keywords associated with the text information are generated.

在本实施例中,可以根据上述文本信息生成与该文本信息相关联的多个关键词,这些关键词可以是与上述产品具有相似名称,或者同样类别,或者相同属性,或者具有相同或相似特征的产品的一些文本形式的描述词汇。In this embodiment, multiple keywords associated with the text information may be generated based on the above text information, and these keywords may have similar names, or the same category, or the same attributes, or have the same or similar characteristics as the above products A descriptive vocabulary of some text form of the product.

例如,某电脑厂商想要获取为其新推出的电脑推荐的广告图像素材时,可以首先输入一些文本信息,这些文本信息可以是电脑的名称,或者是电脑的配置信息,或者是对电脑的新特色的描述的信息等等。如,假设该电脑采用了某种先进的技术改善了显示屏显示效果,则输入的文本信息可以是该先进技术的名称,也可以是显示屏改善后的效果的描述等等。接着,可以根据这些文本信息生成一些关键词,这些关键词可以是其它类的电脑产品的一些描述词汇,如,其它具有类似配置的电脑的名称,或者其它采用类似技术的电脑的简介关键词汇等等。可以理解,上述关键词可以是任意与该文本信息相关联的词汇,本公开对上述关键词的具体形式和内容以及数量方面不限定。For example, when a computer manufacturer wants to obtain the recommended advertising image material for its newly launched computer, it can first input some text information. Features, descriptions, information, etc. For example, assuming that the computer adopts some advanced technology to improve the display effect of the display screen, the input text information may be the name of the advanced technology, or a description of the improved effect of the display screen, etc. Then, some keywords can be generated according to the text information, and these keywords can be some descriptive vocabulary of other types of computer products, such as the names of other computers with similar configurations, or other key words for the introduction of computers with similar technologies, etc. Wait. It can be understood that the above-mentioned keywords may be any vocabulary associated with the text information, and the present disclosure does not limit the specific form, content and quantity of the above-mentioned keywords.

在一种实现方式中,可以从预存的相关词汇集合中获取与上述文本信息关联的相关词汇集合,并将与该文本信息关联的相关词汇集合中包括的词语作为上述关键词。In an implementation manner, a related vocabulary set associated with the text information may be obtained from a pre-stored related vocabulary set, and words included in the related vocabulary set associated with the text information may be used as the keywords.

具体来说,首先,可以预先创建并存储多个相关词汇集合,每个相关词汇集合中包含多个相关的词汇,这些词汇可以在某属性方面具有相关联的关系。如,同一个相关词汇集合中的词汇可以是用于描述相同或相似属性事物的词汇。由于每个词汇可能与多个属性方面有关联,因此,不同的相关词汇集合中也可以包括相同的词汇,换而言之,同一个词汇可能对应于多个相关词汇集合。例如,相关词汇集合A可以包括各种用于描述颜色的词汇,其中可以包括“红色”,相关词汇集合B可以包括各种用于描述服饰的词汇,其中也可以包括“红色”。Specifically, firstly, a plurality of related vocabulary sets may be created and stored in advance, and each related vocabulary set includes a plurality of related words, and these words may have an associated relationship in a certain attribute. For example, the words in the same set of related words may be words used to describe things with the same or similar attributes. Since each vocabulary may be associated with multiple attribute aspects, different related vocabulary sets may also include the same vocabulary, in other words, the same vocabulary may correspond to multiple related vocabulary sets. For example, the related vocabulary set A may include various words used to describe colors, which may include "red", and the related vocabulary set B may include various words used to describe clothing, which may also include "red".

接着,可以从预存的相关词汇集合中获取与上述文本信息关联的相关词汇集合。其中,与上述文本信息关联的相关词汇集合可以是包含从上述文本信息中分解出来的至少一个重点词汇的集合。具体来说,可以先对上述文本信息进行语义解析和语句分解,以分解出一个或多个该文本信息包含的重点词汇,然后,从预存的相关词汇集合中查找包含至少一个上述重点词汇的集合,作为与上述文本信息关联的相关词汇集合。Next, a related vocabulary set associated with the above text information may be obtained from a pre-stored related vocabulary set. Wherein, the related vocabulary set associated with the above text information may be a set including at least one key vocabulary decomposed from the above text information. Specifically, the above-mentioned text information can be semantically analyzed and sentence-decomposed first to decompose one or more key words contained in the text information, and then, a set containing at least one of the above-mentioned key words can be searched from the pre-stored set of related words , as a collection of related words associated with the above text information.

例如,假设,上述文本信息的重点词汇包括A、B、C,预存的相关词汇集合a包括词汇L、O、J、H、G、A,相关词汇集合b包括词汇H、B、R、F、C,相关词汇集合c包括词汇I、T、D、X、D、E、F,相关词汇集合d包括词汇Q、P、N。则由上可知,相关词汇集合a中包括文本信息的重点词汇A,相关词汇集合b中包括文本信息的重点词汇B、C,相关词汇集合c和d中不包括文本信息的重点词汇,因此,可以确定相关词汇集合a和相关词汇集合b为与上述文本信息关联的相关词汇集合。For example, assume that the key vocabulary of the above text information includes A, B, C, the pre-stored related vocabulary set a includes vocabulary L, O, J, H, G, A, and the related vocabulary set b includes vocabulary H, B, R, F , C, the related vocabulary set c includes vocabulary I, T, D, X, D, E, F, and the related vocabulary set d includes vocabulary Q, P, N. It can be seen from the above that the relevant vocabulary set a includes the key vocabulary A of the text information, the relevant vocabulary set b includes the key words B and C of the text information, and the relevant vocabulary sets c and d do not include the key words of the text information. Therefore, It may be determined that the related vocabulary set a and the related vocabulary set b are related vocabulary sets associated with the above text information.

最后,将与文本信息关联的所有相关词汇集合的并集中包括的所有词语作为与上述文本信息关联的关键词。Finally, all words included in the union of all related vocabulary sets associated with the text information are used as keywords associated with the above text information.

在另一种实现方式中,还可以获取预先训练的主题模型,将上述文本信息输入该主题模型,以生成主题词语作为关键词。In another implementation manner, a pre-trained topic model may also be obtained, and the above text information is input into the topic model to generate topic words as keywords.

具体来说,首先,可以获取预先训练好的主题模型,可以从本地存储的数据中获取该主题模型,也可以从其它终端或者服务器获取该主题模型。该主题模型可以是通过本设备训练的,也可以是其它设备训练的。本公开对该主题模型的来源方面不限定。Specifically, first, a pre-trained topic model may be obtained, the topic model may be obtained from locally stored data, or the topic model may be obtained from other terminals or servers. The topic model may be trained by this device or by other devices. This disclosure is not limited in terms of the source of the subject model.

在本实施例中,主题模型可以通过如下方式训练:获取一些文本语句样本,以及获取这些文本语句样本对应的主题词语。然后,采用这些文本语句样本以及对应的主题词语,训练得到主题模型。例如,可以将这些文本语句样本输入待训练的主题模型,计算所得到特征值,将得到的特征值与这些文本语句样本对应的主题词语的特征值进行比较,以得到差值函数,并对待训练主题模型的参数进行调整,重复上述训练过程,直至得到的差值函数收敛为止。In this embodiment, the topic model can be trained by acquiring some text sentence samples and obtaining topic words corresponding to these text sentence samples. Then, the topic model is trained by using these text sentence samples and corresponding topic words. For example, these text sentence samples can be input into the topic model to be trained, the obtained feature values can be calculated, and the obtained feature values can be compared with the feature values of the topic words corresponding to these text sentence samples to obtain a difference function, and to be trained The parameters of the topic model are adjusted, and the above training process is repeated until the obtained difference function converges.

接着,可以将上述文本信息输入该训练好的主题模型,该主题模型对上述文本信息进行处理分析,得到对应的主题词语的特征值,并生成对应的主题词语作为关键词。Next, the above-mentioned text information can be input into the trained topic model, and the topic model can process and analyze the above-mentioned text information, obtain the feature value of the corresponding topic words, and generate the corresponding topic words as keywords.

在步骤103中,根据上述关键词获取推荐的广告图像素材。In step 103, the recommended advertisement image material is obtained according to the above keywords.

在本实施例中,可以根据上述关键词获取一些与上述关键词相关的广告图像素材,作为推荐的广告图像素材。与上述关键词相关的广告图像素材可以是网络中曾经使用过的与上述至少预定个数的关键词相关的广告图像素材。In this embodiment, some advertisement image materials related to the above keywords may be obtained according to the above keywords as recommended advertisement image materials. The advertisement image material related to the above-mentioned keywords may be the advertisement image material related to the above-mentioned at least predetermined number of keywords once used in the network.

例如,假设上述关键词包括3个关键词,则可以预先设定上述预定个数为1,或者2,或者3。当预定个数为1时,与上述关键词相关的广告图像素材可以是与上述关键词中任意一个,或者任意两个,或者全部关键词相关的广告图像素材。当预定个数为2时,与上述关键词相关的广告图像素材可以是与上述关键词中任意两个,或者全部关键词相关的广告图像素材。当预定个数为3时,与上述关键词相关的广告图像素材可以是与上述关键词中全部关键词相关的广告图像素材。可以理解,本公开对此方面不限定。For example, assuming that the above-mentioned keywords include 3 keywords, the above-mentioned predetermined number may be set as 1, or 2, or 3 in advance. When the predetermined number is 1, the advertisement image material related to the above keywords may be advertisement image materials related to any one, any two, or all of the above keywords. When the predetermined number is 2, the advertisement image materials related to the above keywords may be advertisement image materials related to any two or all of the above keywords. When the predetermined number is 3, the advertisement image materials related to the above keywords may be advertisement image materials related to all the keywords in the above keywords. It is to be understood that the present disclosure is not limited in this respect.

在一种实现方式中,可以采用网页爬虫技术从互联网络中爬取与关键词关联的广告图像素材作为推荐的广告图像素材。In an implementation manner, the advertisement image material associated with the keyword may be crawled from the Internet by using a web crawler technology as the recommended advertisement image material.

例如,具体来说,可以采用网页爬虫技术,按照预定的规则,从互联网络中提取出现上述至少预定个数的关键词的网页。然后,从上述网页中获取广告相关网页,并从广告相关网页中提取对应的广告图像素材,作为推荐的广告图像素材。For example, specifically, a web crawler technology may be used to extract web pages containing at least a predetermined number of keywords from the Internet according to predetermined rules. Then, obtain advertisement-related webpages from the above-mentioned webpages, and extract corresponding advertisement image materials from the advertisement-related webpages as recommended advertisement image materials.

在另一种实现方式中,可以将预先建立的广告图像素材库中与关键词匹配的广告图像素材作为推荐的广告图像素材。In another implementation manner, the advertising image material matching the keywords in the pre-established advertising image material library may be used as the recommended advertising image material.

具体来说,可以根据上述关键词,从预先建立的广告图像素材库中查找与上述至少预定个数的关键词匹配的广告图像素材,并将查找出的广告图像素材作为推荐的广告图像素材。Specifically, it is possible to search for advertising image materials that match at least a predetermined number of keywords from a pre-established advertising image material library according to the keywords, and use the found advertising image materials as recommended advertising image materials.

在本实施例中,可以通过如下方式建立广告图像素材库:首先,可以获取现有的一些质量较高的广告图像素材以及对应的关键词,将这些广告图像素材作为参考广告图像素材。然后,将参考广告图像素材及对应的关键词进行关联地存储,以生成广告图像素材库。并且,还可以继续获取更多质量较高的参考广告图像素材以及对应的关键词,并关联地存储到广告图像素材库中,以对广告图像素材库进行补充。In this embodiment, the advertisement image material library can be established in the following manner: First, some existing high-quality advertisement image materials and corresponding keywords can be obtained, and these advertisement image materials can be used as reference advertisement image materials. Then, the reference advertisement image material and the corresponding keywords are associated and stored to generate an advertisement image material library. In addition, more high-quality reference advertising image materials and corresponding keywords can be obtained continuously, and stored in association in the advertising image material library, so as to supplement the advertising image material library.

在步骤104中,输出推荐的广告图像素材。In step 104, the recommended advertisement image material is output.

在本实施例中,获取到推荐的广告图像素材后,需要将推荐的广告图像素材输出,以供用户参考。一般来说,推荐的广告图像素材的质量可能参差不齐,所以,更希望将质量较好的广告图像素材优先推荐给用户,使用户优先看到质量高的广告图像素材。因此,可以先对推荐的广告图像素材的质量进行评估,并根据评估的结果输出推荐的广告图像素材。In this embodiment, after the recommended advertisement image material is acquired, the recommended advertisement image material needs to be output for the user's reference. Generally speaking, the quality of the recommended advertising image materials may vary. Therefore, it is more desirable to recommend better-quality advertising image materials to users first, so that users can preferentially see high-quality advertising image materials. Therefore, the quality of the recommended advertising image material can be evaluated first, and the recommended advertising image material can be output according to the evaluation result.

具体来说,首先,可以分别将推荐的广告图像素材输入到预先训练的广告图像素材质量评估模型中,以获取推荐的广告图像素材对应的质量级别。如,可以将待评估的广告图像素材输入到预先训练的广告图像素材质量评估模型,广告图像素材质量评估模型可以对待评估的广告图像素材进行特征提取及分类,从而对待评估的广告图像素材进行质量评估。接着,将推荐的广告图像素材按照质量级别从高到低的顺序进行排序。并按照上述排序输出推荐的广告图像素材。Specifically, firstly, the recommended advertising image materials may be respectively input into a pre-trained advertising image material quality assessment model, so as to obtain the quality level corresponding to the recommended advertising image materials. For example, the advertising image material to be evaluated can be input into the pre-trained advertising image material quality evaluation model, and the advertising image material quality evaluation model can perform feature extraction and classification on the advertising image material to be evaluated, so as to evaluate the quality of the advertising image material to be evaluated. Evaluate. Next, the recommended advertising image materials are sorted in descending order of quality level. And output the recommended advertising image materials according to the above sorting.

在本实施例中,可以通过如下方式训练广告图像素材质量评估模型:首先,需要预先采集广告图像素材的样本以及对应的质量级别,用采集到的广告图像素材的样本以及对应的质量级别不断输入待训练模型,并不断调整该模型的参数,直到该模型能够比较准确的评估出广告图像素材的样本的质量级别,从而训练得到广告图像素材质量评估模型。当然,本领域中已知的以及将来可能出现的任何其他训练广告图像素材质量评估模型的方法都可以应用于本公开,本公开对其训练广告图像素材质量评估模型的具体方式方面不限定。In this embodiment, the quality evaluation model of the advertising image material can be trained in the following manner: First, samples of the advertising image material and the corresponding quality levels need to be collected in advance, and the collected samples of the advertising image materials and the corresponding quality levels are used to continuously input The model is to be trained, and the parameters of the model are continuously adjusted until the model can accurately evaluate the quality level of samples of advertising image materials, so as to obtain the quality evaluation model of advertising image materials through training. Certainly, any other method for training the quality assessment model of the advertisement image material known in the art and that may appear in the future can be applied to the present disclosure, and the present disclosure does not limit the specific manner of training the quality assessment model of the advertisement image material.

本公开的上述实施例提供的图像素材的推荐方法,通过获取用户输入的文本信息,生成与该文本信息关联的关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,提高了广告投放的效率及效果。The image material recommendation method provided by the above-mentioned embodiments of the present disclosure obtains the text information input by the user, generates a keyword associated with the text information, obtains the recommended advertisement image material according to the keyword, and outputs the recommended advertisement image material . Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, more reference materials can be provided for the user, and the efficiency and effect of advertisement placement can be improved.

如图2所示,图2根据一示例性实施例示出的另一种图像素材的推荐方法的流程图,该实施例详细描述了生成与文本信息关联的关键词的过程,该方法可以应用于终端设备或者服务器中,包括以下步骤:As shown in FIG. 2, FIG. 2 shows a flow chart of another method for recommending image materials according to an exemplary embodiment. This embodiment describes in detail the process of generating keywords associated with text information. This method can be applied to In the terminal device or server, the following steps are included:

在步骤201中,获取互联网络中的历史搜索记录数据。In step 201, historical search record data in the Internet is obtained.

在本实施例中,可以从互联网络中收集部分或全部用户的历史搜索记录数据,该历史搜索记录数据可以是进行任意搜索时,留下的历史记录数据。In this embodiment, some or all of the user's historical search record data may be collected from the Internet, and the historical search record data may be the historical record data left when any search is performed.

在步骤202中,基于历史搜索记录数据生成多个相关词汇集合。In step 202, a plurality of related vocabulary sets are generated based on historical search record data.

在本实施例中,可以基于上述历史搜索记录数据生成多个相关词汇集合。具体来说,可以采用SimRank算法对历史搜索记录数据进行分类处理,以生成多个相关数据集合。In this embodiment, multiple sets of related words may be generated based on the above historical search record data. Specifically, the SimRank algorithm can be used to classify and process historical search record data to generate multiple related data sets.

其中,SimRank算法是一种基于图的拓扑结构信息来衡量结构上下文中对象相似度的普适算法,直观上的含义是利用已有对象的相似度来推算其它有关联对象的相似度。采用SimRank算法求得相似度的核心思想为:如果两个对象和被相似的对象所引用(即它们有相似的入邻边结构),那么这两个对象也相似。Among them, the SimRank algorithm is a universal algorithm based on graph topology information to measure the similarity of objects in the structural context. The intuitive meaning is to use the similarity of existing objects to calculate the similarity of other related objects. The core idea of using the SimRank algorithm to obtain the similarity is: if two objects are referenced by similar objects (that is, they have similar adjacent edge structures), then the two objects are also similar.

可以理解,还可以采用其它的方式生成多个相关词汇集合,本领域中已知的以及将来可能出现的任何其他基于历史搜索记录数据生成多个相关词汇集合的方法都可以应用于本公开,本公开对此方面不限定。It can be understood that other methods can also be used to generate multiple sets of related words, and any other methods of generating multiple sets of related words based on historical search record data known in the art and that may appear in the future can be applied to the present disclosure. Disclosure is not limited in this regard.

在步骤203中,将生成的多个相关词汇集合进行存储。In step 203, the generated sets of related vocabulary are stored.

在步骤204中,获取用户输入的文本信息。In step 204, the text information input by the user is acquired.

在步骤205中,从预存的相关词汇集合中获取与文本信息关联的相关词汇集合。In step 205, a related vocabulary set associated with the text information is acquired from a pre-stored related vocabulary set.

在步骤206中,将与文本信息关联的相关词汇集合中包括的词语作为关键词。In step 206, the words included in the related vocabulary set associated with the text information are used as keywords.

在步骤207中,根据上述关键词获取推荐的广告图像素材。In step 207, the recommended advertisement image material is acquired according to the above keywords.

在步骤208中,输出推荐的广告图像素材。In step 208, the recommended advertisement image material is output.

需要说明的是,对于与图1实施例中相同的步骤,在上述图2实施例中不再进行赘述,相关内容可参见图1实施例。It should be noted that, for the same steps as those in the embodiment in FIG. 1 , they will not be repeated in the above embodiment in FIG. 2 , and relevant content may refer to the embodiment in FIG. 1 .

本公开的上述实施例提供的图像素材的推荐方法,通过基于获取到的历史搜索记录数据生成多个相关词汇集合,再获取用户输入的文本信息,从预存的相关词汇集合中获取与文本信息关联的相关词汇集合,将与文本信息关联的相关词汇集合中包括的词语作为关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高广告投放的效率及效果。The method for recommending image materials provided by the above-mentioned embodiments of the present disclosure generates a plurality of sets of related words based on the acquired historical search record data, and then acquires the text information input by the user, and obtains the related words associated with the text information from the pre-stored set of related words. A set of related words, using words included in the set of related words associated with the text information as keywords, obtaining recommended advertising image materials according to the keywords, and outputting recommended advertising image materials. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

如图3所示,图3是根据一示例性实施例示出的另一种图像素材的推荐方法的流程图,该实施例详细描述了根据关键词获取推荐的广告图像素材的过程,该方法可以应用于终端设备或者服务器中,包括以下步骤:As shown in FIG. 3, FIG. 3 is a flow chart of another method for recommending image materials according to an exemplary embodiment. This embodiment describes in detail the process of obtaining recommended advertising image materials according to keywords. The method can Applied to a terminal device or a server, including the following steps:

在步骤301中,获取参考广告图像素材以及参考广告图像素材对应的关键词。In step 301, a reference advertisement image material and keywords corresponding to the reference advertisement image material are acquired.

在步骤302中,将参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。In step 302, the reference advertisement image material is stored in association with the corresponding keyword, so as to establish an advertisement image material library.

在本实施例中,首先,可以获取现有的一些质量较高的广告图像素材以及对应的关键词,将这些广告图像素材作为参考广告图像素材。然后,将参考广告图像素材及对应的关键词进行关联地存储,以生成广告图像素材库。并且,还可以继续获取更多质量较高的参考广告图像素材以及对应的关键词,并关联地存储到广告图像素材库中,以对广告图像素材库进行补充。In this embodiment, first, some existing high-quality advertising image materials and corresponding keywords may be obtained, and these advertising image materials may be used as reference advertising image materials. Then, the reference advertisement image material and the corresponding keywords are associated and stored to generate an advertisement image material library. In addition, more high-quality reference advertising image materials and corresponding keywords can be obtained continuously, and stored in association in the advertising image material library, so as to supplement the advertising image material library.

在步骤303中,获取用户输入的文本信息。In step 303, the text information input by the user is obtained.

在步骤304中,生成与该文本信息关联的关键词。In step 304, keywords associated with the text information are generated.

在步骤305中,将预先建立的上述广告图像素材库中与上述关键词匹配的广告图像素材作为推荐的广告图像素材。In step 305, the advertisement image material matching the keyword in the pre-established advertisement image material library is used as the recommended advertisement image material.

在步骤306中,输出推荐的广告图像素材。In step 306, the recommended advertisement image material is output.

需要说明的是,对于与图1实施例中相同的步骤,在上述图3实施例中不再进行赘述,相关内容可参见图1实施例。It should be noted that, for the same steps as those in the embodiment in FIG. 1 , they will not be repeated in the embodiment in FIG. 3 , and relevant content may refer to the embodiment in FIG. 1 .

本公开的上述实施例提供的图像素材的推荐方法,通过获取参考广告图像素材以及参考广告图像素材对应的关键词。将参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。再获取用户输入的文本信息,生成与该文本信息关联的关键词,并将预先建立的上述广告图像素材库中与上述关键词匹配的广告图像素材作为推荐的广告图像素材,输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高了广告投放的效率及效果。The method for recommending an image material provided by the above-mentioned embodiments of the present disclosure obtains a reference advertisement image material and keywords corresponding to the reference advertisement image material. The reference advertisement image material is stored in association with the corresponding keyword, so as to establish an advertisement image material library. Then obtain the text information input by the user, generate keywords associated with the text information, and use the advertisement image material matching the above keyword in the pre-established advertisement image material library as the recommended advertisement image material, and output the recommended advertisement image material. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

应当注意,尽管在附图中以特定顺序描述了本公开方法的操作,但是,这并非要求或者暗示必须按照该特定顺序来执行这些操作,或是必须执行全部所示的操作才能实现期望的结果。相反,流程图中描绘的步骤可以改变执行顺序。附加地或备选地,可以省略某些步骤,将多个步骤合并为一个步骤执行,和/或将一个步骤分解为多个步骤执行。It should be noted that, while operations of the disclosed methods are depicted in the figures in a particular order, this does not require or imply that the operations must be performed in that particular order, or that all illustrated operations must be performed to achieve desirable results. . Conversely, the steps depicted in the flowcharts may be performed in an altered order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step for execution, and/or one step may be decomposed into multiple steps for execution.

与前述图像素材的推荐方法实施例相对应,本公开还提供了图像素材的推荐装置的实施例。Corresponding to the above embodiments of the method for recommending image materials, the present disclosure also provides embodiments of an apparatus for recommending image materials.

如图4所示,图4是本公开根据一示例性实施例示出的一种图像素材的推荐装置框图,该装置包括:第一获取模块401,生成模块402,第二获取模块403以及输出模块404。As shown in Figure 4, Figure 4 is a block diagram of an image material recommendation device according to an exemplary embodiment of the present disclosure, the device includes: a first acquisition module 401, a generation module 402, a second acquisition module 403 and an output module 404.

其中,第一获取模块401,被配置为获取用户输入的文本信息。Wherein, the first obtaining module 401 is configured to obtain text information input by the user.

一般来说,产品的广告可以包括很多素材,例如文本素材、图像素材以及声音素材等。其中,图像素材即为图像形式的素材。在本实施例中,当用户想要获取为某产品推荐的广告图像素材时,需要首先输入和该产品相关的文本信息。例如,输入产品的名称,产品的属性,产品的特征,产品的类别以及产品的简介和描述等等,以便获取为该产品推荐的广告图像素材。因此,上述文本信息为与待投放广告对应的产品相关的一些文本信息。可以理解,本公开对文本信息的具体形式和内容方面不限定。Generally speaking, an advertisement of a product may include many materials, such as text material, image material, and sound material. Wherein, the image material is a material in the form of an image. In this embodiment, when the user wants to obtain the recommended advertising image material for a certain product, he needs to first input text information related to the product. For example, the name of the product, the attribute of the product, the feature of the product, the category of the product, the introduction and description of the product, etc. are input, so as to obtain the recommended advertising image material for the product. Therefore, the above text information is some text information related to the product corresponding to the advertisement to be placed. It can be understood that the present disclosure does not limit the specific form and content of the text information.

生成模块402,被配置为生成与文本信息关联的关键词。The generating module 402 is configured to generate keywords associated with the text information.

在本实施例中,可以根据上述文本信息生成与该文本信息相关联的多个关键词,这些关键词可以是与上述产品具有相似名称,或者同样类别,或者相同属性,或者具有相同或相似特征的产品的一些文本形式的描述词汇。In this embodiment, multiple keywords associated with the text information may be generated based on the above text information, and these keywords may have similar names, or the same category, or the same attributes, or have the same or similar characteristics as the above products Descriptive vocabulary in some text form of the product.

在一种实现方式中,可以从预存的相关词汇集合中获取与上述文本信息关联的相关词汇集合,并将与该文本信息关联的相关词汇集合中包括的词语作为上述关键词。In an implementation manner, a related vocabulary set associated with the text information may be obtained from a pre-stored related vocabulary set, and words included in the related vocabulary set associated with the text information may be used as the keywords.

在另一种实现方式中,还可以获取预先训练的主题模型,将上述文本信息输入该主题模型,以生成主题词语作为关键词。In another implementation manner, a pre-trained topic model may also be obtained, and the above text information is input into the topic model to generate topic words as keywords.

第二获取模块403,被配置为根据上述关键词获取推荐的广告图像素材。The second obtaining module 403 is configured to obtain recommended advertisement image materials according to the above keywords.

在本实施例中,可以根据上述关键词获取一些与上述关键词相关的广告图像素材,作为推荐的广告图像素材。与上述关键词相关的广告图像素材可以是网络中曾经使用过的与上述至少预定个数的关键词相关的广告图像素材。In this embodiment, some advertisement image materials related to the above keywords may be obtained according to the above keywords as recommended advertisement image materials. The advertisement image material related to the above-mentioned keywords may be the advertisement image material related to the above-mentioned at least predetermined number of keywords once used in the network.

在一种实现方式中,可以采用网页爬虫技术从互联网络中爬取与关键词关联的广告图像素材作为推荐的广告图像素材。In an implementation manner, the advertisement image material associated with the keyword may be crawled from the Internet by using a web crawler technology as the recommended advertisement image material.

在另一种实现方式中,可以将预先建立的广告图像素材库中与关键词匹配的广告图像素材作为推荐的广告图像素材。In another implementation manner, the advertising image material matching the keywords in the pre-established advertising image material library may be used as the recommended advertising image material.

输出模块404,被配置为输出推荐的广告图像素材。The output module 404 is configured to output recommended advertisement image materials.

在本实施例中,获取到推荐的广告图像素材后,需要将推荐的广告图像素材输出,以供用户参考。一般来说,推荐的广告图像素材的质量可能参差不齐,所以,更希望将质量较好的广告图像素材优先推荐给用户,使用户最先看到质量高的广告图像素材的质量。因此,可以先对推荐的广告图像素材的质量进行评估,并根据评估的结果输出推荐的广告图像素材。In this embodiment, after the recommended advertisement image material is acquired, the recommended advertisement image material needs to be output for the user's reference. Generally speaking, the quality of the recommended advertising image materials may vary. Therefore, it is more desirable to recommend higher-quality advertising image materials to users first, so that users can see the quality of high-quality advertising image materials first. Therefore, the quality of the recommended advertising image material can be evaluated first, and the recommended advertising image material can be output according to the evaluation result.

具体来说,首先,可以分别将推荐的广告图像素材输入到预先训练的广告图像素材质量评估模型中,以获取推荐的广告图像素材对应的质量级别。如,可以将待评估的广告图像素材输入到预先训练的广告图像素材质量评估模型,广告图像素材质量评估模型可以对待评估的广告图像素材进行特征提取及分类,从而对待评估的广告图像素材进行质量评估。接着,将推荐的广告图像素材按照质量级别从高到低的顺序进行排序。并按照上述排序输出推荐的广告图像素材。Specifically, firstly, the recommended advertising image materials may be respectively input into a pre-trained advertising image material quality assessment model, so as to obtain the quality level corresponding to the recommended advertising image materials. For example, the advertising image material to be evaluated can be input into the pre-trained advertising image material quality evaluation model, and the advertising image material quality evaluation model can perform feature extraction and classification on the advertising image material to be evaluated, so as to evaluate the quality of the advertising image material to be evaluated. Evaluate. Next, the recommended advertising image materials are sorted in descending order of quality level. And output the recommended advertising image materials according to the above sorting.

由上述实施例可见,该实施例可以通过获取用户输入的文本信息,生成与该文本信息关联的关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,提高了广告投放的效率及效果。It can be seen from the above embodiments that this embodiment can generate keywords associated with the text information by acquiring the text information input by the user, acquire recommended advertisement image materials according to the keywords, and output the recommended advertisement image materials. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, more reference materials can be provided for the user, and the efficiency and effect of advertisement placement can be improved.

如图5所示,图5是本公开根据一示例性实施例示出的另一种图像素材的推荐装置框图,该实施例在前述图4所示实施例的基础上,生成模块402可以包括:第一获取子模块501和确定子模块502。As shown in FIG. 5, FIG. 5 is a block diagram of an apparatus for recommending another image material according to an exemplary embodiment of the present disclosure. On the basis of the embodiment shown in FIG. 4, the generating module 402 may include: A first acquiring submodule 501 and a determining submodule 502 .

其中,第一获取子模块501,被配置为从预存的相关词汇集合中获取与上述文本信息关联的相关词汇集合。Wherein, the first acquiring sub-module 501 is configured to acquire a related vocabulary set associated with the above text information from a pre-stored related vocabulary set.

确定子模块502,被配置为将与该文本信息关联的相关词汇集合中包括的词语作为关键词。The determining submodule 502 is configured to use words included in the related vocabulary set associated with the text information as keywords.

在本实施例中,首先,可以预先创建并存储多个相关词汇集合,每个相关词汇集合中包含多个相关的词汇,这些词汇可以在某属性方面具有相关联的关系。由于每个词汇可能与多个属性方面有关联,因此,不同的相关词汇集合中也可以包括相同的词汇,换而言之,同一个词汇可能对应于多个相关词汇集合。In this embodiment, first, a plurality of related vocabulary sets may be created and stored in advance, and each related vocabulary set includes a plurality of related words, and these words may have an associated relationship in a certain attribute. Since each vocabulary may be associated with multiple attribute aspects, different related vocabulary sets may also include the same vocabulary, in other words, the same vocabulary may correspond to multiple related vocabulary sets.

接着,可以从预存的相关词汇集合中获取与上述文本信息关联的相关词汇集合。其中,与上述文本信息关联的相关词汇集合可以是包含从上述文本信息中分解出来的至少一个重点词汇的集合。具体来说,可以先对上述文本信息进行语义解析和语句分解,以分解出一个或多个该文本信息包含的重点词汇,然后,从预存的相关词汇集合中查找包含至少一个上述重点词汇的集合,作为与上述文本信息关联的相关词汇集合。Next, a related vocabulary set associated with the above text information may be obtained from a pre-stored related vocabulary set. Wherein, the related vocabulary set associated with the above text information may be a set including at least one key vocabulary decomposed from the above text information. Specifically, the above-mentioned text information can be semantically analyzed and sentence-decomposed first to decompose one or more key words contained in the text information, and then, a set containing at least one of the above-mentioned key words can be searched from the pre-stored set of related words , as a collection of related words associated with the above text information.

最后,将与文本信息关联的所有相关词汇集合的并集中包括的所有词语作为与上述文本信息关联的关键词。Finally, all words included in the union of all related vocabulary sets associated with the text information are used as keywords associated with the above text information.

由上述实施例可见,该实施例可以获取用户输入的文本信息,从预存的相关词汇集合中获取与上述文本信息关联的相关词汇集合,将与该文本信息关联的相关词汇集合中包括的词语作为关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高广告投放的效率及效果。It can be seen from the above embodiment that this embodiment can acquire the text information input by the user, obtain the related vocabulary set associated with the above text information from the pre-stored related vocabulary set, and use the words included in the related vocabulary set associated with the text information as A keyword, according to which the recommended advertising image material is obtained, and the recommended advertising image material is output. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

如图6所示,图6是本公开根据一示例性实施例示出的另一种图像素材的推荐装置框图,该实施例在前述图4所示实施例的基础上,生成模块402可以包括:第二获取子模块601和生成子模块602。As shown in FIG. 6, FIG. 6 is a block diagram of an apparatus for recommending another image material according to an exemplary embodiment of the present disclosure. On the basis of the embodiment shown in FIG. 4, the generating module 402 may include: The second acquiring sub-module 601 and generating sub-module 602 .

其中,第二获取子模块601,被配置为获取预先训练的主题模型。Wherein, the second obtaining sub-module 601 is configured to obtain a pre-trained topic model.

在本实施例中,可以获取预先训练好的主题模型,可以从本地存储的数据中获取该主题模型,也可以从其它终端或者服务器获取该主题模型。该主题模型可以是通过本设备训练的,也可以是其它设备训练的。本公开对该主题模型的来源方面不限定。In this embodiment, the pre-trained topic model can be obtained, the topic model can be obtained from locally stored data, or the topic model can be obtained from other terminals or servers. The topic model may be trained by this device or by other devices. This disclosure is not limited in terms of the source of the subject model.

在本实施例中,主题模型可以通过如下方式训练:获取一些文本语句样本,以及获取这些文本语句样本对应的主题词语。然后,采用这些文本语句样本以及对应的主题词语,训练得到主题模型。例如,可以将这些文本语句样本输入待训练的主题模型,计算所得到特征值,将得到的特征值与这些文本语句样本对应的主题词语的特征值进行比较,以得到差值函数,并对待训练主题模型的参数进行调整,重复上述训练过程,直至得到的差值函数收敛为止。In this embodiment, the topic model can be trained by acquiring some text sentence samples and obtaining topic words corresponding to these text sentence samples. Then, the topic model is trained by using these text sentence samples and corresponding topic words. For example, these text sentence samples can be input into the topic model to be trained, the obtained eigenvalues can be calculated, and the obtained eigenvalues can be compared with the eigenvalues of the topic words corresponding to these text sentence samples to obtain a difference function, and to be trained The parameters of the topic model are adjusted, and the above training process is repeated until the obtained difference function converges.

生成子模块602,被配置为将上述文本信息输入该主题模型,以生成主题词语作为关键词。The generation sub-module 602 is configured to input the above text information into the topic model to generate topic words as keywords.

在本实施例中,可以将上述文本信息输入该训练好的主题模型,该主题模型对上述文本信息进行处理分析,得到对应的主题词语的特征值,并生成对应的主题词语作为关键词。In this embodiment, the above-mentioned text information can be input into the trained topic model, and the topic model processes and analyzes the above-mentioned text information, obtains the feature values of the corresponding topic words, and generates the corresponding topic words as keywords.

由上述实施例可见,该实施例可以获取用户输入的文本信息,再获取预先训练的主题模型,将上述文本信息输入该主题模型,以生成主题词语作为关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高广告投放的效率及效果。It can be seen from the above embodiment that this embodiment can obtain the text information input by the user, and then obtain the pre-trained topic model, input the above text information into the topic model to generate topic words as keywords, and obtain recommended advertisements based on the keywords image material, and output recommended ad image material. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

如图7所示,图7是本公开根据一示例性实施例示出的另一种图像素材的推荐装置框图,该实施例在前述图4所示实施例的基础上,第二获取模块403可以包括:爬取子模块701。As shown in FIG. 7, FIG. 7 is a block diagram of an apparatus for recommending another image material according to an exemplary embodiment of the present disclosure. On the basis of the embodiment shown in FIG. 4, the second acquisition module 403 can Including: crawling sub-module 701.

其中,爬取子模块701,被配置为采用网页爬虫技术从互联网络中爬取与上述关键词关联的广告图像素材作为推荐的广告图像素材。Wherein, the crawling sub-module 701 is configured to crawl the advertisement image material associated with the above keywords from the Internet as the recommended advertisement image material by using web crawler technology.

在本实施例中,可以采用网页爬虫技术,按照预定的规则,从互联网络中提取出现上述至少预定个数的关键词的网页。然后,从上述网页中获取广告相关网页,并从广告相关网页中提取对应的广告图像素材,作为推荐的广告图像素材。In this embodiment, a web crawler technology may be used to extract web pages containing at least a predetermined number of keywords from the Internet according to predetermined rules. Then, obtain advertisement-related webpages from the above-mentioned webpages, and extract corresponding advertisement image materials from the advertisement-related webpages as recommended advertisement image materials.

由上述实施例可见,该实施例可以通过获取用户输入的文本信息,生成与该文本信息关联的关键词,采用网页爬虫技术从互联网络中爬取与该关键词关联的广告图像素材作为推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,从更丰富的广告图像素材中获取推荐广告图像素材,并向用户推荐广告图像素材,进一步为用户提供更多的参考素材,并提高了广告投放的效率及效果。As can be seen from the above-mentioned embodiment, this embodiment can generate a keyword associated with the text information by obtaining the text information input by the user, and use web crawler technology to crawl the advertisement image material associated with the keyword from the Internet as a recommended Advertisement image material, and output recommended ad image material. Therefore, according to the text information input by the user, the recommended advertising image material can be obtained from more abundant advertising image materials, and the advertising image material can be recommended to the user, further providing more reference materials for the user, and improving the efficiency and effect of advertising placement .

如图8所示,图8是本公开根据一示例性实施例示出的另一种图像素材的推荐装置框图,该实施例在前述图4所示实施例的基础上,第二获取模块403可以包括:推荐子模块801。As shown in FIG. 8, FIG. 8 is a block diagram of an apparatus for recommending another image material according to an exemplary embodiment of the present disclosure. On the basis of the embodiment shown in FIG. 4, the second acquisition module 403 may Including: recommendation sub-module 801 .

其中,推荐子模块801,被配置为将预先建立的广告图像素材库中与上述关键词匹配的广告图像素材作为推荐的广告图像素材。Wherein, the recommendation sub-module 801 is configured to use the advertisement image material matching the above keywords in the pre-established advertisement image material library as the recommended advertisement image material.

在本实施例中,可以根据上述关键词,从预先建立的广告图像素材库中查找与上述至少预定个数的关键词匹配的广告图像素材,并将查找出的广告图像素材作为推荐的广告图像素材。In this embodiment, according to the above keywords, it is possible to search for advertisement image materials that match at least a predetermined number of keywords from a pre-established advertisement image material database, and use the searched advertisement image materials as recommended advertisement images material.

由上述实施例可见,该实施例可以通过获取用户输入的文本信息,生成与该文本信息关联的关键词,将预先建立的广告图像素材库中与上述关键词匹配的广告图像素材作为推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,获取较高质量的推荐广告图像素材,并向用户推荐广告图像素材,为用户提供更多的参考素材,并进一步提高了广告投放的效率及效果。It can be seen from the above-mentioned embodiment that this embodiment can generate keywords associated with the text information by acquiring the text information input by the user, and use the advertisement image material matching the above-mentioned keywords in the pre-established advertisement image material library as the recommended advertisement image material, and output recommended ad image material. Therefore, according to the text information input by the user, higher-quality recommended advertisement image materials can be obtained, and the advertisement image materials can be recommended to the user, providing more reference materials for the user, and further improving the efficiency and effect of advertisement placement.

如图9所示,图9是本公开根据一示例性实施例示出的另一种图像素材的推荐装置框图,该实施例在前述图5所示实施例的基础上,该装置还可以进一步包括:第三获取模块405,集合生成模块406和存储模块407。As shown in FIG. 9, FIG. 9 is a block diagram of an apparatus for recommending another image material according to an exemplary embodiment of the present disclosure. On the basis of the embodiment shown in FIG. 5, the apparatus may further include : the third acquisition module 405, the set generation module 406 and the storage module 407.

其中,第三获取模块405,被配置为获取互联网络中的历史搜索记录数据。Wherein, the third obtaining module 405 is configured to obtain historical search record data in the Internet.

在本实施例中,可以从互联网络中收集部分或全部用户的历史搜索记录数据,该历史搜索记录数据可以是进行任意搜索时,留下的历史记录数据。In this embodiment, some or all of the user's historical search record data may be collected from the Internet, and the historical search record data may be the historical record data left when any search is performed.

集合生成模块406,被配置为基于历史搜索记录数据生成多个相关词汇集合。The set generating module 406 is configured to generate multiple sets of related words based on historical search record data.

在本实施例中,可以基于上述历史搜索记录数据生成多个相关词汇集合。具体来说,可以采用SimRank算法对历史搜索记录数据进行分类处理,以生成多个相关数据集合。In this embodiment, multiple sets of related words may be generated based on the above historical search record data. Specifically, the SimRank algorithm can be used to classify and process historical search record data to generate multiple related data sets.

存储模块407,被配置为将生成的多个相关词汇集合进行存储。The storage module 407 is configured to store the generated sets of related vocabulary.

由上述实施例可见,该实施例可以通过基于获取到的历史搜索记录数据生成多个相关词汇集合,再获取用户输入的文本信息,从预存的相关词汇集合中获取与文本信息关联的相关词汇集合,将与文本信息关联的相关词汇集合中包括的词语作为关键词,根据该关键词获取推荐的广告图像素材,并输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高广告投放的效率及效果。It can be seen from the above-mentioned embodiment that this embodiment can generate a plurality of related vocabulary sets based on the obtained historical search record data, and then obtain the text information input by the user, and obtain the related vocabulary sets associated with the text information from the pre-stored related vocabulary sets , using the words included in the relevant vocabulary set associated with the text information as keywords, obtaining recommended advertising image materials according to the keywords, and outputting the recommended advertising image materials. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

如图10所示,图10是本公开根据一示例性实施例示出的另一种图像素材的推荐装置框图,该实施例在前述图9所示实施例的基础上,集合生成模块406可以包括:处理子模块1001。As shown in FIG. 10, FIG. 10 is a block diagram of an apparatus for recommending another image material according to an exemplary embodiment of the present disclosure. In this embodiment, on the basis of the embodiment shown in FIG. 9, the set generating module 406 may include : Processing sub-module 1001.

其中,处理子模块1001,被配置为采用SimRank算法对上述历史搜索记录数据进行分类处理,以生成多个相关数据集合。Wherein, the processing sub-module 1001 is configured to classify and process the above-mentioned historical search record data by using the SimRank algorithm, so as to generate multiple related data sets.

由上述实施例可见,该实施例通过采用SimRank算法生成多个相关数据集合,提高了生成相关数据集合的效率。It can be seen from the above embodiment that this embodiment improves the efficiency of generating related data sets by using the SimRank algorithm to generate multiple related data sets.

如图11所示,图11是本公开根据一示例性实施例示出的另一种图像素材的推荐装置框图,该实施例在前述图6所示实施例的基础上,该装置还可以进一步包括:第四获取模块408和建立模块409。As shown in FIG. 11, FIG. 11 is a block diagram of another device for recommending image materials according to an exemplary embodiment of the present disclosure. On the basis of the embodiment shown in FIG. 6, the device may further include : the fourth obtaining module 408 and the establishing module 409.

其中,第四获取模块408,被配置为获取参考广告图像素材以及参考广告图像素材对应的关键词。Wherein, the fourth acquiring module 408 is configured to acquire the reference advertisement image material and the keywords corresponding to the reference advertisement image material.

建立模块409,被配置为将参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。The establishment module 409 is configured to store the reference advertisement image material in association with the corresponding keyword, so as to establish an advertisement image material library.

在本实施例中,首先,可以获取现有的一些质量较高的广告图像素材以及对应的关键词,将这些广告图像素材作为参考广告图像素材。然后,将参考广告图像素材及对应的关键词进行关联地存储,以生成广告图像素材库。并且,还可以继续获取更多质量较高的参考广告图像素材以及对应的关键词,并关联地存储到广告图像素材库中,以对广告图像素材库进行补充。In this embodiment, first, some existing high-quality advertising image materials and corresponding keywords may be acquired, and these advertising image materials may be used as reference advertising image materials. Then, the reference advertisement image material and the corresponding keywords are associated and stored to generate an advertisement image material library. In addition, more high-quality reference advertising image materials and corresponding keywords can be obtained continuously, and stored in association in the advertising image material library, so as to supplement the advertising image material library.

由上述实施例可见,该实施例通过获取参考广告图像素材以及参考广告图像素材对应的关键词,将参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。再获取用户输入的文本信息,生成与该文本信息关联的关键词,并将预先建立的上述广告图像素材库中与上述关键词匹配的广告图像素材作为推荐的广告图像素材,输出推荐的广告图像素材。从而能够根据用户输入的文本信息,向用户推荐广告图像素材,为用户提供更多的参考素材,有助于提高了广告投放的效率及效果。It can be seen from the above embodiment that in this embodiment, the reference advertisement image material and the keywords corresponding to the reference advertisement image material are obtained, and the reference advertisement image material and the corresponding keyword are associated and stored, so as to establish an advertisement image material library. Then obtain the text information input by the user, generate keywords associated with the text information, and use the advertisement image material matching the above keyword in the pre-established advertisement image material library as the recommended advertisement image material, and output the recommended advertisement image material. Therefore, according to the text information input by the user, the advertisement image material can be recommended to the user, and more reference materials can be provided for the user, which helps to improve the efficiency and effect of advertisement placement.

如图12所示,图12是本公开根据一示例性实施例示出的另一种图像素材的推荐装置框图,该实施例在前述图4所示实施例的基础上,输出模块404,可以包括:输入子模块1201,排序子模块1202和输出子模块1203。As shown in FIG. 12, FIG. 12 is a block diagram of another image material recommendation device according to an exemplary embodiment of the present disclosure. On the basis of the embodiment shown in FIG. 4, the output module 404 may include : an input sub-module 1201 , a sorting sub-module 1202 and an output sub-module 1203 .

其中,输入子模块1201,被配置为分别将推荐的广告图像素材输入到预先训练的广告图像素材质量评估模型,以获取推荐的广告图像素材对应的质量级别。Wherein, the input sub-module 1201 is configured to respectively input the recommended advertisement image material into the pre-trained advertisement image material quality assessment model, so as to obtain the quality level corresponding to the recommended advertisement image material.

在本实施例中,可以分别将推荐的广告图像素材输入到预先训练的广告图像素材质量评估模型中,以获取推荐的广告图像素材对应的质量级别。如,可以将待评估的广告图像素材输入到预先训练的广告图像素材质量评估模型,广告图像素材质量评估模型可以对待评估的广告图像素材进行特征提取及分类,从而对待评估的广告图像素材进行质量评估。In this embodiment, the recommended advertising image materials may be respectively input into a pre-trained advertising image material quality assessment model, so as to obtain the quality level corresponding to the recommended advertising image materials. For example, the advertising image material to be evaluated can be input into the pre-trained advertising image material quality evaluation model, and the advertising image material quality evaluation model can perform feature extraction and classification on the advertising image material to be evaluated, so as to evaluate the quality of the advertising image material to be evaluated. Evaluate.

在本实施例中,可以通过如下方式训练广告图像素材质量评估模型:首先,需要预先采集广告图像素材的样本以及对应的质量级别,用采集到的广告图像素材的样本以及对应的质量级别不断输入待训练模型,并不断调整该模型的参数,直到该模型能够比较准确的评估出广告图像素材的样本的质量级别,从而训练得到广告图像素材质量评估模型。当然,本领域中已知的以及将来可能出现的任何其他训练广告图像素材质量评估模型的方法都可以应用于本公开,本公开对其训练广告图像素材质量评估模型的具体方式方面不限定。In this embodiment, the quality evaluation model of the advertising image material can be trained in the following manner: First, samples of the advertising image material and the corresponding quality levels need to be collected in advance, and the collected samples of the advertising image materials and the corresponding quality levels are used to continuously input The model is to be trained, and the parameters of the model are continuously adjusted until the model can accurately evaluate the quality level of samples of advertising image materials, so as to obtain the quality evaluation model of advertising image materials through training. Certainly, any other method for training the quality assessment model of the advertisement image material known in the art and that may appear in the future can be applied to the present disclosure, and the present disclosure does not limit the specific manner of training the quality assessment model of the advertisement image material.

排序子模块1202,被配置为将推荐的广告图像素材按照质量级别从高到低的顺序进行排序。The sorting sub-module 1202 is configured to sort the recommended advertisement image materials in descending order of quality levels.

输出子模块1203,被配置为按照上述排序输出推荐的广告图像素材。The output sub-module 1203 is configured to output the recommended advertisement image material according to the above order.

由上述实施例可见,该实施例按照质量级别从高到低的顺序对获取到的推荐的广告图像素材进行排序,并按照上述排序输出推荐的广告图像素材。从而可以使用户优先看到质量较高的广告图像素材,并为用户提供更多的参考素材,进一步提高了广告投放的效率及效果。It can be seen from the above-mentioned embodiment that this embodiment sorts the obtained recommended advertisement image materials in descending order of quality levels, and outputs the recommended advertisement image materials according to the above-mentioned sorting. Therefore, the user can preferentially see the high-quality advertisement image material, and provide the user with more reference materials, further improving the efficiency and effect of advertisement placement.

应当理解,上述装置可以预先设置在终端或者服务器中,也可以通过下载等方式而加载到终端或者服务器中。上述装置中的相应模块可以与终端或者服务器中的模块相互配合以实现图像素材的推荐方案。It should be understood that the above devices may be pre-set in the terminal or the server, and may also be loaded into the terminal or the server by means of downloading or the like. The corresponding modules in the above device can cooperate with the modules in the terminal or the server to realize the recommendation scheme of the image material.

对于装置实施例而言,由于其基本对应于方法实施例,所以相关之处参见方法实施例的部分说明即可。以上所描述的装置实施例仅仅是示意性的,其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本公开方案的目的。本领域普通技术人员在不付出创造性劳动的情况下,即可以理解并实施。As for the device embodiment, since it basically corresponds to the method embodiment, for related parts, please refer to the part description of the method embodiment. The device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. It can be understood and implemented by those skilled in the art without creative effort.

相应的,本公开还提供一种图像素材的推荐装置,该装置包括有处理器;用于存储处理器可执行指令的存储器;其中,该处理器被配置为:Correspondingly, the present disclosure also provides a device for recommending image materials, the device includes a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to:

获取用户输入的文本信息;Obtain the text information entered by the user;

生成与所述文本信息关联的关键词;generating keywords associated with the text information;

根据所述关键词获取推荐的广告图像素材;Acquiring recommended advertisement image materials according to the keywords;

输出所述推荐的广告图像素材。Outputting the recommended advertisement image material.

图13是根据一示例性实施例示出的一种图像素材的推荐装置9900的一结构示意图。例如,装置9900可以是移动电话,计算机,数字广播终端,消息收发设备,游戏控制台,平板设备,医疗设备,健身设备,个人数字助理等。Fig. 13 is a schematic structural diagram of an apparatus 9900 for recommending image materials according to an exemplary embodiment. For example, the apparatus 9900 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, and the like.

参照图13,装置9900可以包括以下一个或多个组件:处理组件9902,存储器9904,电源组件9906,多媒体组件9908,音频组件9910,输入/输出(I/O)的接口9912,传感器组件9914,以及通信组件9916。13, the device 9900 may include one or more of the following components: a processing component 9902, a memory 9904, a power supply component 9906, a multimedia component 9908, an audio component 9910, an input/output (I/O) interface 9912, a sensor component 9914, And the communication component 9916.

处理组件9902通常控制装置9900的整体操作,诸如与显示,电话呼叫,数据通信,相机操作和记录操作相关联的操作。处理元件9902可以包括一个或多个处理器9920来执行指令,以完成上述的方法的全部或部分步骤。此外,处理组件9902可以包括一个或多个模块,便于处理组件9902和其他组件之间的交互。例如,处理组件9902可以包括多媒体模块,以方便多媒体组件9908和处理组件9902之间的交互。The processing component 9902 generally controls the overall operations of the device 9900, such as those associated with display, telephone calls, data communications, camera operations, and recording operations. The processing element 9902 may include one or more processors 9920 to execute instructions to complete all or part of the steps of the above method. Additionally, processing component 9902 may include one or more modules that facilitate interaction between processing component 9902 and other components. For example, the processing component 9902 can include a multimedia module to facilitate interaction between the multimedia component 9908 and the processing component 9902.

存储器9904被配置为存储各种类型的数据以支持在装置9900的操作。这些数据的示例包括用于在装置9900上操作的任何应用程序或方法的指令,联系人数据,电话簿数据,消息,图片,视频等。存储器9904可以由任何类型的易失性或非易失性存储设备或者它们的组合实现,如静态随机存取存储器(SRAM),电可擦除可编程只读存储器(EEPROM),可擦除可编程只读存储器(EPROM),可编程只读存储器(PROM),只读存储器(ROM),磁存储器,快闪存储器,磁盘或光盘。The memory 9904 is configured to store various types of data to support operations at the device 9900 . Examples of such data include instructions for any application or method operating on the device 9900, contact data, phonebook data, messages, pictures, videos, etc. The memory 9904 can be implemented by any type of volatile or non-volatile memory device or their combination, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic or Optical Disk.

电源组件9906为装置9900的各种组件提供电力。电源组件9906可以包括电源管理系统,一个或多个电源,及其他与为装置9900生成、管理和分配电力相关联的组件。The power supply component 9906 provides power to the various components of the device 9900. Power components 9906 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for device 9900.

多媒体组件9908包括在所述装置9900和用户之间的提供一个输出接口的屏幕。在一些实施例中,屏幕可以包括液晶显示器(LCD)和触摸面板(TP)。如果屏幕包括触摸面板,屏幕可以被实现为触摸屏,以接收来自用户的输入信号。触摸面板包括一个或多个触摸传感器以感测触摸、滑动和触摸面板上的手势。所述触摸传感器可以不仅感测触摸或滑动动作的边界,而且还检测与所述触摸或滑动操作相关的持续时间和压力。在一些实施例中,多媒体组件9908包括一个前置摄像头和/或后置摄像头。当装置9900处于操作模式,如拍摄模式或视频模式时,前置摄像头和/或后置摄像头可以接收外部的多媒体数据。每个前置摄像头和后置摄像头可以是一个固定的光学透镜系统或具有焦距和光学变焦能力。The multimedia component 9908 includes a screen that provides an output interface between the device 9900 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may not only sense a boundary of a touch or swipe action, but also detect duration and pressure associated with the touch or swipe action. In some embodiments, multimedia component 9908 includes a front camera and/or rear camera. When the device 9900 is in an operation mode, such as a shooting mode or a video mode, the front camera and/or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have focal length and optical zoom capability.

音频组件9910被配置为输出和/或输入音频信号。例如,音频组件9910包括一个麦克风(MIC),当装置9900处于操作模式,如呼叫模式、记录模式和语音识别模式时,麦克风被配置为接收外部音频信号。所接收的音频信号可以被进一步存储在存储器9904或经由通信组件9916发送。在一些实施例中,音频组件9910还包括一个扬声器,用于输出音频信号。The audio component 9910 is configured to output and/or input audio signals. For example, the audio component 9910 includes a microphone (MIC), which is configured to receive an external audio signal when the device 9900 is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode. Received audio signals may be further stored in memory 9904 or transmitted via communication component 9916. In some embodiments, the audio component 9910 also includes a speaker for outputting audio signals.

I/O接口9912为处理组件9902和外围接口模块之间提供接口,上述外围接口模块可以是键盘,点击轮,按钮等。这些按钮可包括但不限于:主页按钮、音量按钮、启动按钮和锁定按钮。The I/O interface 9912 provides an interface between the processing component 9902 and a peripheral interface module, which may be a keyboard, a click wheel, a button, and the like. These buttons may include, but are not limited to: a home button, volume buttons, start button, and lock button.

传感器组件9914包括一个或多个传感器,用于为装置9900提供各个方面的状态评估。例如,传感器组件9914可以检测到装置9900的打开/关闭状态,组件的相对定位,例如所述组件为装置9900的显示器和小键盘,传感器组件9914还可以检测装置9900或装置9900一个组件的位置改变,用户与装置9900接触的存在或不存在,装置9900方位或加速/减速和装置9900的温度变化。传感器组件9914可以包括接近传感器,被配置用来在没有任何的物理接触时检测附近物体的存在。传感器组件9914还可以包括光传感器,如CMOS或CCD图像传感器,用于在成像应用中使用。在一些实施例中,该传感器组件9914还可以包括加速度传感器,陀螺仪传感器,磁传感器,压力传感器,微波传感器或温度传感器。Sensor assembly 9914 includes one or more sensors for providing status assessments of various aspects of device 9900. For example, the sensor assembly 9914 can detect the open/closed state of the device 9900, the relative positioning of components such as the display and keypad of the device 9900, the sensor assembly 9914 can also detect a change in the position of the device 9900 or a component of the device 9900 , the presence or absence of user contact with the device 9900, the device 9900 orientation or acceleration/deceleration and the temperature change of the device 9900. Sensor assembly 9914 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact. The sensor assembly 9914 may also include light sensors, such as CMOS or CCD image sensors, for use in imaging applications. In some embodiments, the sensor assembly 9914 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, a microwave sensor or a temperature sensor.

通信组件9916被配置为便于装置9900和其他设备之间有线或无线方式的通信。装置9900可以接入基于通信标准的无线网络,如WiFi,2G或3G,或它们的组合。在一个示例性实施例中,通信组件9916经由广播信道接收来自外部广播管理系统的广播信号或广播相关信息。在一个示例性实施例中,所述通信组件9916还包括近场通信(NFC)模块,以促进短程通信。例如,在NFC模块可基于射频识别(RFID)技术,红外数据协会(IrDA)技术,超宽带(UWB)技术,蓝牙(BT)技术和其他技术来实现。The communication component 9916 is configured to facilitate wired or wireless communication between the apparatus 9900 and other devices. The device 9900 can access wireless networks based on communication standards, such as WiFi, 2G or 3G, or a combination thereof. In one exemplary embodiment, the communication component 9916 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communications component 9916 also includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wide Band (UWB) technology, Bluetooth (BT) technology and other technologies.

在示例性实施例中,装置9900可以被一个或多个应用专用集成电路(ASIC)、数字信号处理器(DSP)、数字信号处理设备(DSPD)、可编程逻辑器件(PLD)、现场可编程门阵列(FPGA)、控制器、微控制器、微处理器或其他电子元件实现,用于执行上述方法。In an exemplary embodiment, apparatus 9900 may be programmed by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable A gate array (FPGA), controller, microcontroller, microprocessor or other electronic component implementation for performing the methods described above.

在示例性实施例中,还提供了一种包括指令的非临时性计算机可读存储介质,例如包括指令的存储器9904,上述指令可由装置9900的处理器9920执行以完成上述方法。例如,所述非临时性计算机可读存储介质可以是ROM、随机存取存储器(RAM)、CD-ROM、磁带、软盘和光数据存储设备等。In an exemplary embodiment, there is also provided a non-transitory computer-readable storage medium including instructions, such as a memory 9904 including instructions, which can be executed by the processor 9920 of the device 9900 to implement the above method. For example, the non-transitory computer readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like.

本领域技术人员在考虑说明书及实践这里公开的发明后,将容易想到本公开的其它实施方案。本申请旨在涵盖本公开的任何变型、用途或者适应性变化,这些变型、用途或者适应性变化遵循本公开的一般性原理并包括本公开未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的,本公开的真正范围和精神由下面的权利要求指出。Other embodiments of the present disclosure will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the present disclosure, and these modifications, uses or adaptations follow the general principles of the present disclosure and include common knowledge or conventional technical means in the technical field not disclosed in the present disclosure . The specification and examples are to be considered exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

应当理解的是,本公开并不局限于上面已经描述并在附图中示出的精确结构,并且可以在不脱离其范围进行各种修改和改变。本公开的范围仅由所附的权利要求来限制。It should be understood that the present disclosure is not limited to the precise constructions which have been described above and shown in the drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims (19)

1.一种图像素材的推荐方法,其特征在于,所述方法包括:1. A recommended method for image material, characterized in that the method comprises: 获取用户输入的文本信息;Obtain the text information entered by the user; 生成与所述文本信息关联的关键词;generating keywords associated with the text information; 根据所述关键词获取推荐的广告图像素材;Acquiring recommended advertisement image materials according to the keywords; 输出所述推荐的广告图像素材。Outputting the recommended advertisement image material. 2.根据权利要求1所述的方法,其特征在于,所述生成与所述文本信息关联的关键词,包括:2. The method according to claim 1, wherein said generating keywords associated with said text information comprises: 从预存的相关词汇集合中获取与所述文本信息关联的相关词汇集合;Acquiring a related vocabulary set associated with the text information from a pre-stored related vocabulary set; 将与所述文本信息关联的相关词汇集合中包括的词语作为所述关键词。The words included in the related vocabulary set associated with the text information are used as the keywords. 3.根据权利要求1所述的方法,其特征在于,所述生成与所述文本信息关联的关键词,包括:3. The method according to claim 1, wherein said generating keywords associated with said text information comprises: 获取预先训练的主题模型;Get a pre-trained topic model; 将所述文本信息输入所述主题模型,以生成主题词语作为所述关键词。The text information is input into the topic model to generate topic words as the keywords. 4.根据权利要求1所述的方法,其特征在于,所述根据所述关键词获取推荐的广告图像素材,包括:4. The method according to claim 1, wherein said acquiring recommended advertisement image material according to said keywords comprises: 采用网页爬虫技术从互联网络中爬取与所述关键词关联的广告图像素材作为推荐的广告图像素材。Advertisement image materials associated with the keywords are crawled from the Internet by web crawler technology as recommended advertisement image materials. 5.根据权利要求1所述的方法,其特征在于,所述根据所述关键词获取推荐的广告图像素材,包括:5. The method according to claim 1, wherein said acquiring recommended advertisement image material according to said keyword comprises: 将预先建立的广告图像素材库中与所述关键词匹配的广告图像素材作为推荐的广告图像素材。Advertisement image materials matching the keywords in the pre-established advertisement image material library are used as recommended advertisement image materials. 6.根据权利要求2所述的方法,其特征在于,所述方法还包括:6. The method according to claim 2, characterized in that the method further comprises: 获取互联网络中的历史搜索记录数据;Obtain historical search record data in the Internet; 基于所述历史搜索记录数据生成多个相关词汇集合;generating a plurality of related vocabulary sets based on the historical search record data; 将生成的多个相关词汇集合进行存储。Store the generated sets of related vocabulary. 7.根据权利要求6所述的方法,其特征在于,所述基于所述历史搜索记录数据生成多个相关词汇集合,包括:7. The method according to claim 6, wherein said generating a plurality of related vocabulary sets based on said historical search record data comprises: 采用SimRank算法对所述历史搜索记录数据进行分类处理,以生成多个相关数据集合。The SimRank algorithm is used to classify the historical search record data to generate a plurality of related data sets. 8.根据权利要求5所述的方法,其特征在于,所述方法还包括:8. The method according to claim 5, further comprising: 获取参考广告图像素材以及所述参考广告图像素材对应的关键词;Obtaining a reference advertisement image material and keywords corresponding to the reference advertisement image material; 将所述参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。The reference advertisement image material is stored in association with corresponding keywords to establish an advertisement image material library. 9.根据权利要求1所述的方法,其特征在于,所述输出所述推荐的广告图像素材,包括:9. The method according to claim 1, wherein said outputting said recommended advertisement image material comprises: 分别将所述推荐的广告图像素材输入到预先训练的广告图像素材质量评估模型,以获取所述推荐的广告图像素材对应的质量级别;respectively inputting the recommended advertising image material into a pre-trained advertising image material quality evaluation model to obtain the quality level corresponding to the recommended advertising image material; 将所述推荐的广告图像素材按照质量级别从高到低的顺序进行排序;sorting the recommended advertising image materials in order of quality level from high to low; 按照所述排序输出所述推荐的广告图像素材。Outputting the recommended advertisement image material according to the ranking. 10.一种图像素材的推荐装置,其特征在于,所述装置包括:10. A device for recommending image materials, characterized in that the device comprises: 第一获取模块,被配置为获取用户输入的文本信息;The first obtaining module is configured to obtain the text information input by the user; 生成模块,被配置为生成与所述文本信息关联的关键词;a generation module configured to generate keywords associated with the text information; 第二获取模块,被配置为根据所述关键词获取推荐的广告图像素材;The second acquiring module is configured to acquire the recommended advertisement image material according to the keywords; 输出模块,被配置为输出所述推荐的广告图像素材。An output module configured to output the recommended advertising image material. 11.根据权利要求10所述的装置,其特征在于,所述生成模块包括:11. The device according to claim 10, wherein the generating module comprises: 第一获取子模块,被配置为从预存的相关词汇集合中获取与所述文本信息关联的相关词汇集合;The first acquisition submodule is configured to acquire a related vocabulary set associated with the text information from a pre-stored related vocabulary set; 确定子模块,被配置为将与所述文本信息关联的相关词汇集合中包括的词语作为所述关键词。The determining submodule is configured to use the words included in the related vocabulary set associated with the text information as the keywords. 12.根据权利要求10所述的装置,其特征在于,所述生成模块包括:12. The device according to claim 10, wherein the generating module comprises: 第二获取子模块,被配置为获取预先训练的主题模型;The second obtaining submodule is configured to obtain a pre-trained topic model; 生成子模块,被配置为将所述文本信息输入所述主题模型,以生成主题词语作为所述关键词。A generating submodule configured to input the text information into the topic model to generate topic words as the keywords. 13.根据权利要求10所述的装置,其特征在于,所述第二获取模块包括:13. The device according to claim 10, wherein the second acquiring module comprises: 爬取子模块,被配置为采用网页爬虫技术从互联网络中爬取与所述关键词关联的广告图像素材作为推荐的广告图像素材。The crawling sub-module is configured to crawl the advertising image material associated with the keyword from the Internet as a recommended advertising image material by using web crawler technology. 14.根据权利要求10所述的装置,其特征在于,所述第二获取模块包括:14. The device according to claim 10, wherein the second acquiring module comprises: 推荐子模块,被配置为将预先建立的广告图像素材库中与所述关键词匹配的广告图像素材作为推荐的广告图像素材。The recommending sub-module is configured to use the advertising image material matching the keyword in the pre-established advertising image material library as the recommended advertising image material. 15.根据权利要求11所述的装置,其特征在于,所述装置还包括:15. The device according to claim 11, further comprising: 第三获取模块,被配置为获取互联网络中的历史搜索记录数据;The third obtaining module is configured to obtain historical search record data in the Internet; 集合生成模块,被配置为基于所述历史搜索记录数据生成多个相关词汇集合;A collection generating module configured to generate a plurality of related vocabulary collections based on the historical search record data; 存储模块,被配置为将生成的多个相关词汇集合进行存储。The storage module is configured to store the generated sets of related vocabulary. 16.根据权利要求15所述的装置,其特征在于,所述集合生成模块包括:16. The device according to claim 15, wherein the set generating module comprises: 处理子模块,被配置为采用SimRank算法对所述历史搜索记录数据进行分类处理,以生成多个相关数据集合。The processing sub-module is configured to use the SimRank algorithm to classify and process the historical search record data to generate a plurality of related data sets. 17.根据权利要求14所述的装置,其特征在于,所述装置还包括:17. The device according to claim 14, further comprising: 第四获取模块,被配置为获取参考广告图像素材以及所述参考广告图像素材对应的关键词;The fourth obtaining module is configured to obtain a reference advertisement image material and keywords corresponding to the reference advertisement image material; 建立模块,被配置为将所述参考广告图像素材与对应的关键词进行关联地存储,以建立广告图像素材库。The establishment module is configured to store the reference advertisement image material in association with corresponding keywords, so as to establish an advertisement image material library. 18.根据权利要求10所述的装置,其特征在于,所述输出模块包括:18. The device according to claim 10, wherein the output module comprises: 输入子模块,被配置为分别将所述推荐的广告图像素材输入到预先训练的广告图像素材质量评估模型,以获取所述推荐的广告图像素材对应的质量级别;The input sub-module is configured to respectively input the recommended advertising image material into the pre-trained advertising image material quality evaluation model, so as to obtain the quality level corresponding to the recommended advertising image material; 排序子模块,被配置为将所述推荐的广告图像素材按照质量级别从高到低的顺序进行排序;The sorting submodule is configured to sort the recommended advertising image materials in order of quality level from high to low; 输出子模块,被配置为按照所述排序输出所述推荐的广告图像素材。The output sub-module is configured to output the recommended advertising image material according to the ranking. 19.一种图像素材的推荐装置,其特征在于,包括:19. A device for recommending image materials, comprising: 处理器;processor; 用于存储处理器可执行指令的存储器;memory for storing processor-executable instructions; 其中,所述处理器被配置为:Wherein, the processor is configured as: 获取用户输入的文本信息;Obtain the text information entered by the user; 生成与所述文本信息关联的关键词;generating keywords associated with the text information; 根据所述关键词获取推荐的广告图像素材;Acquiring recommended advertisement image materials according to the keywords; 输出所述推荐的广告图像素材。Outputting the recommended advertisement image material.
CN201610676576.9A 2016-08-16 2016-08-16 Image material recommendation method and device Pending CN106355429A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610676576.9A CN106355429A (en) 2016-08-16 2016-08-16 Image material recommendation method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610676576.9A CN106355429A (en) 2016-08-16 2016-08-16 Image material recommendation method and device

Publications (1)

Publication Number Publication Date
CN106355429A true CN106355429A (en) 2017-01-25

Family

ID=57844914

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610676576.9A Pending CN106355429A (en) 2016-08-16 2016-08-16 Image material recommendation method and device

Country Status (1)

Country Link
CN (1) CN106355429A (en)

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107562963A (en) * 2017-10-12 2018-01-09 杭州群核信息技术有限公司 A kind of method and apparatus screened house ornamentation design and render figure
WO2018145590A1 (en) * 2017-02-13 2018-08-16 腾讯科技(深圳)有限公司 Emoticon picture recommendation method and device, server cluster and storage medium
CN108415961A (en) * 2018-02-06 2018-08-17 厦门集微科技有限公司 A kind of advertising pictures recommendation method and device
CN108733779A (en) * 2018-05-04 2018-11-02 百度在线网络技术(北京)有限公司 The method and apparatus of text figure
CN109783798A (en) * 2018-12-12 2019-05-21 平安科技(深圳)有限公司 Method, apparatus, terminal and the storage medium of text information addition picture
CN109783749A (en) * 2018-12-10 2019-05-21 深圳变设龙信息科技有限公司 A kind of Material for design intelligent recommendation method, apparatus and terminal device
CN110020200A (en) * 2019-03-15 2019-07-16 微梦创科网络科技(中国)有限公司 A kind of personalized recommendation method and system based on history material
CN110471721A (en) * 2018-05-10 2019-11-19 北京京东尚科信息技术有限公司 Page display method and system, electronic equipment and storage medium
CN110858914A (en) * 2018-08-23 2020-03-03 北京优酷科技有限公司 Video material recommendation method and device
CN111259180A (en) * 2020-01-14 2020-06-09 广州视源电子科技股份有限公司 Image pushing method and device, electronic equipment and storage medium
CN111598638A (en) * 2019-02-21 2020-08-28 北京沃东天骏信息技术有限公司 Click rate determination method, device and equipment
CN111790158A (en) * 2019-10-31 2020-10-20 厦门雅基软件有限公司 Game scene editing method and device, electronic equipment and readable storage medium
CN112256907A (en) * 2020-11-17 2021-01-22 携程计算机技术(上海)有限公司 Travel strategy editing method, system, equipment and storage medium based on photo library
CN112651332A (en) * 2020-12-24 2021-04-13 携程旅游信息技术(上海)有限公司 Scene facility identification method, system, equipment and storage medium based on photo library
CN113220917A (en) * 2020-02-06 2021-08-06 阿里巴巴集团控股有限公司 Background map recommendation method, device and storage medium
CN110580299B (en) * 2018-06-08 2023-11-07 北京京东尚科信息技术有限公司 Method, system, equipment and storage medium for generating matching diagram of recommended language of object
CN118485123A (en) * 2024-06-11 2024-08-13 广州逸虎网络科技有限公司 A model training and information recommendation method, device, storage medium and equipment
CN119228443A (en) * 2024-11-28 2024-12-31 长春电子科技学院 A method and system for generating short video advertisements

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6611348B1 (en) * 1999-07-30 2003-08-26 Banta Corporation System and method for communication over a TCP/IP network with an appletalk network for publishing and printing services
US6738155B1 (en) * 1999-07-30 2004-05-18 Banta Corporation System and method of providing publishing and printing services via a communications network
CN101859423A (en) * 2009-04-10 2010-10-13 索尼公司 Server apparatus, method of producing advertisement information, and program
CN103699625A (en) * 2013-12-20 2014-04-02 北京百度网讯科技有限公司 Method and device for retrieving based on keyword
CN104391942A (en) * 2014-11-25 2015-03-04 中国科学院自动化研究所 Short text characteristic expanding method based on semantic atlas

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6611348B1 (en) * 1999-07-30 2003-08-26 Banta Corporation System and method for communication over a TCP/IP network with an appletalk network for publishing and printing services
US6738155B1 (en) * 1999-07-30 2004-05-18 Banta Corporation System and method of providing publishing and printing services via a communications network
CN101859423A (en) * 2009-04-10 2010-10-13 索尼公司 Server apparatus, method of producing advertisement information, and program
CN103699625A (en) * 2013-12-20 2014-04-02 北京百度网讯科技有限公司 Method and device for retrieving based on keyword
CN104391942A (en) * 2014-11-25 2015-03-04 中国科学院自动化研究所 Short text characteristic expanding method based on semantic atlas

Cited By (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10824669B2 (en) 2017-02-13 2020-11-03 Tencent Technology (Shenzhen) Company Limited Sticker recommendation method and apparatus, server cluster, and storage medium
WO2018145590A1 (en) * 2017-02-13 2018-08-16 腾讯科技(深圳)有限公司 Emoticon picture recommendation method and device, server cluster and storage medium
CN107562963A (en) * 2017-10-12 2018-01-09 杭州群核信息技术有限公司 A kind of method and apparatus screened house ornamentation design and render figure
CN107562963B (en) * 2017-10-12 2021-04-20 杭州群核信息技术有限公司 Method and device for screening home decoration design rendering graph
CN108415961A (en) * 2018-02-06 2018-08-17 厦门集微科技有限公司 A kind of advertising pictures recommendation method and device
CN108733779A (en) * 2018-05-04 2018-11-02 百度在线网络技术(北京)有限公司 The method and apparatus of text figure
CN108733779B (en) * 2018-05-04 2022-10-04 百度在线网络技术(北京)有限公司 Text matching method and device
CN110471721B (en) * 2018-05-10 2024-06-18 北京京东尚科信息技术有限公司 Page display method and system, electronic equipment and storage medium
CN110471721A (en) * 2018-05-10 2019-11-19 北京京东尚科信息技术有限公司 Page display method and system, electronic equipment and storage medium
CN110580299B (en) * 2018-06-08 2023-11-07 北京京东尚科信息技术有限公司 Method, system, equipment and storage medium for generating matching diagram of recommended language of object
CN110858914A (en) * 2018-08-23 2020-03-03 北京优酷科技有限公司 Video material recommendation method and device
CN110858914B (en) * 2018-08-23 2021-11-26 阿里巴巴(中国)有限公司 Video material recommendation method and device
CN109783749A (en) * 2018-12-10 2019-05-21 深圳变设龙信息科技有限公司 A kind of Material for design intelligent recommendation method, apparatus and terminal device
CN109783798A (en) * 2018-12-12 2019-05-21 平安科技(深圳)有限公司 Method, apparatus, terminal and the storage medium of text information addition picture
CN111598638A (en) * 2019-02-21 2020-08-28 北京沃东天骏信息技术有限公司 Click rate determination method, device and equipment
CN111598638B (en) * 2019-02-21 2023-11-07 北京沃东天骏信息技术有限公司 Click rate determination method, device and equipment
CN110020200A (en) * 2019-03-15 2019-07-16 微梦创科网络科技(中国)有限公司 A kind of personalized recommendation method and system based on history material
CN111790158A (en) * 2019-10-31 2020-10-20 厦门雅基软件有限公司 Game scene editing method and device, electronic equipment and readable storage medium
CN111259180A (en) * 2020-01-14 2020-06-09 广州视源电子科技股份有限公司 Image pushing method and device, electronic equipment and storage medium
CN111259180B (en) * 2020-01-14 2024-04-19 广州视源电子科技股份有限公司 Image pushing method, device, electronic equipment and storage medium
CN113220917A (en) * 2020-02-06 2021-08-06 阿里巴巴集团控股有限公司 Background map recommendation method, device and storage medium
CN113220917B (en) * 2020-02-06 2022-04-12 阿里巴巴集团控股有限公司 Background map recommendation method, device and storage medium
CN112256907B (en) * 2020-11-17 2024-04-16 携程计算机技术(上海)有限公司 Travel attack editing method, system, equipment and storage medium based on photo library
CN112256907A (en) * 2020-11-17 2021-01-22 携程计算机技术(上海)有限公司 Travel strategy editing method, system, equipment and storage medium based on photo library
CN112651332A (en) * 2020-12-24 2021-04-13 携程旅游信息技术(上海)有限公司 Scene facility identification method, system, equipment and storage medium based on photo library
CN118485123A (en) * 2024-06-11 2024-08-13 广州逸虎网络科技有限公司 A model training and information recommendation method, device, storage medium and equipment
CN119228443A (en) * 2024-11-28 2024-12-31 长春电子科技学院 A method and system for generating short video advertisements
CN119228443B (en) * 2024-11-28 2025-02-07 长春电子科技学院 A method and system for generating short video advertisements

Similar Documents

Publication Publication Date Title
CN106355429A (en) Image material recommendation method and device
CN108038102B (en) Recommended method, device, terminal and storage medium for facial expression images
CN106600331A (en) Method and apparatus for running advertisements
CN108717403B (en) Processing method and device for processing
CN110399548A (en) A kind of search processing method, device, electronic equipment and storage medium
CN111046210B (en) Information recommendation method, device and electronic equipment
CN106777016B (en) Method and device for information recommendation based on instant messaging
CN112508612B (en) Method for training advertisement creative generation model and generating advertisement creative and related device
WO2021147949A1 (en) Video recommendation method and apparatus
CN110309324A (en) A search method and related device
CN105373580A (en) Method and device for displaying subjects
CN106294933A (en) Method, device, terminal and the server that 3D prints
CN106815291B (en) Search result item display method and device and search result item display device
CN110929122B (en) Data processing method and device for data processing
TWI782137B (en) Method and device for generating and displaying data object information
CN114676308A (en) Method, device, electronic device, storage medium and product for search term recommendation
CN107622074A (en) A kind of data processing method, device and computing device
CN109948043A (en) A kind of information recommendation method and device
CN111831132B (en) Information recommendation method, device and electronic device
CN106682163A (en) Article information recommendation method and device and equipment
CN112784151A (en) Method and related device for determining recommendation information
CN106960026B (en) Search method, search engine and electronic equipment
CN108241438B (en) Input method, input device and input device
CN106951541A (en) Cooking information recommendation, acquisition methods and device
CN116521962A (en) Public opinion data mining method, system, electronic equipment and storage medium

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20170125

RJ01 Rejection of invention patent application after publication