CN114611004A - Recommendation method and device of content information, computer equipment and storage medium - Google Patents
Recommendation method and device of content information, computer equipment and storage medium Download PDFInfo
- Publication number
- CN114611004A CN114611004A CN202210285462.7A CN202210285462A CN114611004A CN 114611004 A CN114611004 A CN 114611004A CN 202210285462 A CN202210285462 A CN 202210285462A CN 114611004 A CN114611004 A CN 114611004A
- Authority
- CN
- China
- Prior art keywords
- algorithm
- user
- recommendation algorithm
- recommendation
- target
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Transfer Between Computers (AREA)
Abstract
本申请涉及内容推荐领域，特别是涉及到一种内容信息的推荐方法、装置、计算机设备及存储介质，所述方法包括：获取用户浏览内容信息的行为数据；根据所述行为数据进行分类，并根据分类结果生成用户画像标签以及内容数据标签；获取不同策略的第一推荐算法和第二推荐算法；接收用户对第一推荐算法、所述第二推荐算法的选定指令，根据所述选定指令确定目标推荐算法；根据所述用户画像标签、所述内容数据标签与所述目标推荐算法向用户推送目标内容信息。本申请综合考虑用户主动意图以及用户画像与内容数据，能够有效解决推荐系统中的马太效应问题，提高内容信息个性化推送的精准性与多样性。
The present application relates to the field of content recommendation, and in particular to a method, device, computer equipment and storage medium for recommending content information. The method includes: acquiring behavior data of a user browsing content information; classifying according to the behavior data, and Generate user portrait tags and content data tags according to the classification results; obtain the first recommendation algorithm and the second recommendation algorithm of different strategies; receive the user's selection instruction for the first recommendation algorithm and the second recommendation algorithm, The instruction determines a target recommendation algorithm; pushes target content information to the user according to the user portrait tag, the content data tag and the target recommendation algorithm. This application comprehensively considers the user's active intention, user portrait and content data, can effectively solve the problem of the Matthew effect in the recommendation system, and improve the accuracy and diversity of personalized push of content information.
Description
Technical Field
The present application relates to the field of content recommendation, and in particular, to a method and an apparatus for recommending content information, a computer device, and a storage medium.
Background
At present, basic information, browsing content, browsing duration, product preference and the like of a user are recorded in a background by a mainstream audio and video online playing platform, a news platform, a content propagation platform, an e-commerce shopping platform and the like on the market, the recorded information is used as an input data source of a personalized recommendation engine, and prediction calculation is carried out according to a preset recommendation algorithm, so that other possibly interesting contents are recommended to the user, and the purpose of reserving more user resources is achieved.
Personalized recommendation helps users to filter out much information according to the interests and needs of the users to a large extent, but also causes a phenomenon: the users see and hear what the platform calculates based on a preset logic algorithm is what the users see and hear, which obviously does not play a role of a recommendation system, but influences the shopping experience of the users, and the accuracy and diversity of the personalized push of the content information are lower.
Disclosure of Invention
The present application mainly aims to provide a method and an apparatus for recommending content information, a computer device, and a storage medium, and aims to solve the problem that the accuracy and diversity of personalized push of content information in a recommendation system are low.
In order to achieve the above object, the present application provides a method for recommending content information, the method comprising:
acquiring behavior data of content information browsed by a user;
classifying according to the behavior data, and generating a user portrait label and a content data label according to a classification result;
acquiring a first recommendation algorithm and a second recommendation algorithm of different strategies;
receiving a selected instruction of a user for a first recommendation algorithm and a second recommendation algorithm, and determining a target recommendation algorithm according to the selected instruction;
and pushing target content information to the user according to the user portrait label, the content data label and the target recommendation algorithm.
Further, after obtaining the first recommendation algorithm and the second recommendation algorithm of different policies, the method further includes:
configuring the first recommendation algorithm according to the user portrait label, wherein the first recommendation algorithm comprises a plurality of first sub-algorithms;
and configuring the second recommendation algorithm according to the content data tag, wherein the second recommendation algorithm comprises a plurality of second sub-algorithms.
Further, the receiving a selected instruction of the user for the first recommendation algorithm and the second recommendation algorithm, and determining a target recommendation algorithm according to the selected instruction includes:
receiving a selected instruction of a user for the first sub-algorithm and the second sub-algorithm, and obtaining a target sub-algorithm according to the selected instruction;
and determining a target recommendation algorithm according to the target sub-algorithm.
Further, the receiving a selected instruction of the user for the first sub-algorithm and the second sub-algorithm, and obtaining a target sub-algorithm according to the selected instruction includes:
displaying a first sub-algorithm contained in the first recommendation algorithm and a second sub-algorithm contained in the second recommendation algorithm to a user according to a preset subdivision label;
receiving a selected instruction of the user for the subdivision label, wherein the selected instruction is generated based on clicking operation behavior data;
and screening and determining a target sub-algorithm from the first sub-algorithm and the second sub-algorithm according to the clicking operation behavior data.
Further, the pushing target content information to the user according to the user portrait label, the content data label and the target recommendation algorithm includes:
acquiring recommendation proportions of a plurality of strategies in the target recommendation algorithm;
determining target content information according to the recommendation proportion, the user portrait label and the content data label, wherein the target content information comprises a plurality of different amounts of content information;
a corresponding amount of content information is pushed to the user.
Further, the obtaining of the first recommendation algorithm and the second recommendation algorithm of different policies includes:
acquiring a plurality of algorithms to be selected with different strategies from an algorithm library;
acquiring the historical pushing accuracy of the algorithm to be selected;
sorting the algorithms to be selected according to the historical pushing accuracy;
and obtaining the sorted pre-preset number of algorithms to be selected as a first recommendation algorithm and a second recommendation algorithm.
Further, after the target content information is pushed to the user according to the user portrait label, the content data label and the target recommendation algorithm, the method further comprises:
receiving a switching instruction for switching the target recommendation algorithm by a user;
acquiring a first target recommendation algorithm according to the switching instruction;
and determining first recommended content information according to the first target recommendation algorithm, the user portrait label and the content data label, and displaying the first recommended content information in a pre-browsing mode.
The present application further provides a device for recommending content information, including:
the behavior data module is used for acquiring behavior data of the content information browsed by the user;
the data classification module is used for classifying according to the behavior data and generating a user portrait label and a content data label according to a classification result;
the algorithm obtaining module is used for obtaining a first recommendation algorithm and a second recommendation algorithm of different strategies;
the algorithm screening module is used for receiving a selected instruction of a user for the first recommendation algorithm and the second recommendation algorithm and determining a target recommendation algorithm according to the selected instruction;
and the information pushing module is used for pushing target content information to the user according to the user portrait label, the content data label and the target recommendation algorithm.
The application also provides a computer device, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the recommendation method of any content information when executing the computer program.
The present application also provides a computer-readable storage medium having a computer program stored thereon, where the computer program is executed by a processor to implement any one of the above methods for recommending content information.
The method comprises the steps of firstly obtaining behavior data of content information browsed by a user, then analyzing the behavior data, modifying a recommendation algorithm according to the behavior data and recommendation preference of the user, pushing more accurate target content for the user, specifically, classifying according to the behavior data, generating user portrait labels and content data labels according to classification results, then obtaining a first recommendation algorithm and a second recommendation algorithm of different strategies, opening preset recommendation algorithms (namely the first recommendation algorithm and the second recommendation algorithm) to the user for selection by a platform, enabling the user to select any one or more algorithms from the open recommendation algorithms as target recommendation algorithms, and receiving the first recommendation algorithm, the second recommendation algorithm, the third recommendation algorithm, the fourth recommendation algorithm, the fifth recommendation algorithm and the fifth recommendation algorithm from the user, The recommendation algorithm selected by the user is defined as a target recommendation algorithm, target content information is pushed to the user according to the user portrait label, the content data label and the target recommendation algorithm, different recommendation algorithms are configured for different users, the initiative intention of the user and the user portrait and the content data are comprehensively considered, the Martian effect problem in a recommendation system can be effectively solved, and the accuracy and diversity of personalized pushing of the content information are improved.
Drawings
Fig. 1 is a schematic flowchart illustrating an embodiment of a method for recommending content information according to the present application;
FIG. 2 is a schematic flowchart of an embodiment of the present application after acquiring a first recommendation algorithm and a second recommendation algorithm of different policies;
FIG. 3 is a schematic flowchart of an embodiment of receiving a user's selected instruction for a first recommendation algorithm and a second recommendation algorithm, and determining a target recommendation algorithm according to the selected instruction;
FIG. 4 is a schematic flowchart of an embodiment of the present application, where a user selects an instruction for the first sub-algorithm and the second sub-algorithm, and a target sub-algorithm is obtained according to the selected instruction;
FIG. 5 is a schematic flow chart illustrating an embodiment of pushing target content information to a user according to the user portrait tag, the content data tag, and the target recommendation algorithm in the present application;
FIG. 6 is a schematic flowchart of an embodiment of a first recommendation algorithm and a second recommendation algorithm for obtaining different policies according to the present application;
FIG. 7 is a schematic flow chart illustrating an embodiment of the present application after target content information is pushed to a user according to the user portrait tag, the content data tag, and the target recommendation algorithm;
FIG. 8 is a schematic structural diagram of an embodiment of a content information recommendation device according to the present application;
FIG. 9 is a block diagram illustrating a computer device according to an embodiment of the present disclosure.
The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
Referring to fig. 1, an embodiment of the present application provides a method for recommending content information, where the method for recommending content information includes steps S101-S105, and details of each step of the method for recommending content information are described as follows.
S101, behavior data of the user browsing content information is obtained.
The embodiment is applied to a content recommendation and pushing scene, in which a platform generally provides content, is configured with a recommendation system, and pushes the content to a user according to respective data of the user, and the user can browse the corresponding content on the platform, wherein the platform comprises an audio and video online playing platform, a news platform, a content propagation platform and an e-commerce shopping platform. The embodiment can be applied to the audio and video online playing platform, the news, the content propagation platform and the e-commerce shopping platform by developing a recommendation system, wherein the audio and video online playing platform, the news, the content propagation platform and the e-commerce shopping platform can record basic information, browsing content, browsing duration, product preference and the like of a user, namely record data of the user when the user browses the content on the platform, define the data as behavior data of the user browsing content information, namely acquire the behavior data of the user browsing content information, analyze the behavior data, modify a recommendation algorithm according to the behavior data and the recommendation preference of the user, and push more accurate target content for the user.
And S102, classifying according to the behavior data, and generating a user portrait label and a content data label according to a classification result.
In this embodiment, after behavior data of content information browsed by a user is acquired, behavior data of different content information browsed by the user, including browsing duration, browsing frequency, comment times and the like, is acquired, then, classification is performed according to the behavior data, a user portrait tag and a content data tag are generated according to a classification result, the behavior data of the content information browsed by the user is converted into a user portrait in a big data statistics manner, the user portrait includes a plurality of user portrait tags, and meanwhile, the user portrait is converted into a content data tag according to the content information browsed by the user, that is, the frequency, the duration and the like of different types of content information browsed by the user are acquired from the behavior data of the content information browsed by the user, and then, a corresponding content information tag is extracted as the content data tag.
S103, acquiring a first recommendation algorithm and a second recommendation algorithm of different strategies.
In this embodiment, after the behavior data are classified and the user portrait label and the content data label are generated according to the classification result, a first recommendation algorithm and a second recommendation algorithm with different strategies are obtained, that is, recommendation algorithms with a plurality of different strategies are configured in the recommendation system, more accurate content information can be pushed to the user according to the user portrait label and the content data label of the user through different strategies, and specifically, the first recommendation algorithm and the second recommendation algorithm with different strategies are obtained from the algorithm pool.
S104, receiving a selected instruction of the user on the first recommendation algorithm and the second recommendation algorithm, and determining a target recommendation algorithm according to the selected instruction.
In this embodiment, after a first recommendation algorithm and a second recommendation algorithm of different strategies are obtained, a push strategy for pushing content information to an open user is performed at the same time, that is, a preset recommendation algorithm (that is, the first recommendation algorithm and the second recommendation algorithm) is opened to a user by a platform for selection, instead of adopting a set of large and uniform recommendation algorithm and strategy for all users, the user may select any one or more algorithms from the open recommendation algorithm as a target recommendation algorithm, specifically, receive a selection instruction of the user for the first recommendation algorithm and the second recommendation algorithm, determine the recommendation algorithm selected by the user according to the selection instruction, and define the recommendation algorithm selected by the user as the target recommendation algorithm. The first recommendation algorithm and the second recommendation algorithm comprise a recommendation algorithm based on demographics, a recommendation algorithm based on content, a recommendation algorithm based on collaborative filtering, a mixed recommendation algorithm and a non-recommendation algorithm.
And S105, pushing target content information to the user according to the user portrait label, the content data label and the target recommendation algorithm.
In this embodiment, after receiving a selection instruction of a user for a first recommendation algorithm and a second recommendation algorithm, determining a target recommendation algorithm according to the selection instruction, that is, determining what policy the user selects as its recommended content information based on, and then pushing the target content information to the user according to the user portrait tag, the content data tag, and the target recommendation algorithm, specifically, in a recommendation algorithm model, replacing an initial algorithm of the recommendation algorithm model with the target recommendation algorithm, then using the user portrait tag and the content data tag as input data of the recommendation algorithm model, and after the input data is input to the recommendation algorithm model, calculating the recommended content with the target recommendation algorithm obtained by a plurality of recommendation algorithms selected by the user to obtain the target content information, and then pushing the target content information to the user, different recommendation algorithms are configured for different users, the initiative intention of the users and the user image and content data are comprehensively considered, the problem of the Martian effect in a recommendation system can be effectively solved, and the accuracy and diversity of the personalized push of the content information are improved. In the embodiment, an open recall strategy is removed, namely, a preset recommendation algorithm is opened to the user for selection in the background, instead of adopting a set of large and uniform recommendation algorithm and strategy for all users, so that the problem of the Martian effect in the recommendation system can be effectively solved, namely, hot things can be seen by more people, hot things can become more hot, cold things are more cold, and under different use scenes, the user can automatically determine the categories needing to be recommended, for example, when watching videos, besides videos with the same content metadata, preference contents of other users, historical preference contents of other users and the like can be selected.
The embodiment provides a method for pushing content information to a user by comprehensively considering user initiative intention and user image and content data, the method comprises the steps of firstly obtaining behavior data of content information browsed by the user, then analyzing the behavior data, modifying a recommendation algorithm according to the behavior data and recommendation preference of the user, pushing more accurate target content for the user, specifically, classifying according to the behavior data, generating user image labels and content data labels according to classification results, then obtaining a first recommendation algorithm and a second recommendation algorithm of different strategies, opening preset recommendation algorithms (namely the first recommendation algorithm and the second recommendation algorithm) to the user for selection through a platform, enabling the user to select any one or more algorithms from the opened recommendation algorithms as target recommendation algorithms, and receiving the first recommendation algorithm, the second recommendation algorithm, the third recommendation algorithm, the fourth recommendation algorithm, the fifth recommendation algorithm and the fifth algorithm of the user, The recommendation algorithm selected by the user is defined as a target recommendation algorithm, target content information is pushed to the user according to the user portrait label, the content data label and the target recommendation algorithm, different recommendation algorithms are configured for different users, the initiative intention of the user and the user portrait and the content data are comprehensively considered, the Martian effect problem in a recommendation system can be effectively solved, and the accuracy and diversity of personalized pushing of the content information are improved.
In one embodiment, as shown in fig. 2, after obtaining the first recommendation algorithm and the second recommendation algorithm of different policies, the method further includes steps S201 to S202:
s201, configuring the first recommendation algorithm according to the user portrait label, wherein the first recommendation algorithm comprises a plurality of first sub-algorithms;
s202, configuring the second recommendation algorithm according to the content data labels, wherein the second recommendation algorithm comprises a plurality of second sub-algorithms.
In this embodiment, after a first recommendation algorithm and a second recommendation algorithm of different strategies are obtained, the first recommendation algorithm includes a demographic-based recommendation algorithm, that is, a user portrait-based recommendation algorithm, and then the first recommendation algorithm is configured according to the user portrait tags, where the first recommendation algorithm includes a plurality of first sub-algorithms, and the first recommendation algorithm is configured to include a plurality of user portrait tags, and each user portrait tag is configured to be a corresponding first sub-algorithm, that is, the first sub-algorithm includes a user portrait tag-based recommendation algorithm, … …, and a user portrait tag-based N recommendation algorithm. For a second recommendation algorithm, the second recommendation algorithm includes a content-based recommendation algorithm, and then the second recommendation algorithm is configured according to the content data tags, the second recommendation algorithm includes a number of second sub-algorithms, the second recommendation algorithm is configured to include a plurality of content data tags, and each content data tag is configured as a corresponding second sub-algorithm, that is, the second sub-algorithms include a content data tag-based recommendation algorithm, a content data tag-based second recommendation algorithm, a content data tag-based third recommendation algorithm, … …, and a content data tag-based N recommendation algorithm. By configuring sub-algorithms of the first recommendation algorithm and the second recommendation algorithm, the extensibility of the algorithms is improved, the comprehensiveness and diversity of the algorithms are improved, and the accuracy of content information pushing is improved.
In one embodiment, as shown in fig. 3, the receiving a selected instruction of the user for the first recommendation algorithm and the second recommendation algorithm, and determining a target recommendation algorithm according to the selected instruction includes steps S301 to S302:
s301, receiving a selected instruction of a user for the first sub-algorithm and the second sub-algorithm, and obtaining a target sub-algorithm according to the selected instruction;
s302, determining a target recommendation algorithm according to the target sub-algorithm.
In the embodiment, in the process of receiving a selected instruction of a user for a first recommendation algorithm and a second recommendation algorithm and determining a target recommendation algorithm according to the selected instruction, the selected instruction of the user for the first sub-algorithm and the second sub-algorithm is received, the target sub-algorithm is obtained according to the selected instruction, the target recommendation algorithm is determined according to the target sub-algorithm, namely, an algorithm selected by the user from the configured first sub-algorithm and the configured second sub-algorithm is selected as the target sub-algorithm, then fusion is performed based on the selected target sub-algorithm to obtain the target recommendation algorithm, the recommendation algorithm of the user can be accurately configured according to different user portrait tags and content data tags, different recommendation algorithms are configured for different users, the initiative intention of the user and the user portrait and content data are comprehensively considered, and the problem of the martagon effect in a recommendation system can be effectively solved, the accuracy and diversity of the personalized pushing of the content information are improved.
Further, the first recommendation algorithm further comprises a collaborative filtering-based recommendation algorithm, and the sub-algorithms based on the collaborative filtering recommendation algorithm comprise a user-based collaborative filtering recommendation algorithm, an article-based collaborative filtering recommendation algorithm and a model-based collaborative filtering recommendation algorithm; the second recommendation algorithm further comprises a hybrid recommendation algorithm, and sub-algorithms of the hybrid recommendation algorithm comprise a weighted hybrid recommendation algorithm, a switched hybrid recommendation algorithm, a partitioned hybrid recommendation algorithm, and a layered hybrid recommendation algorithm.
In one embodiment, as shown in fig. 4, the receiving a user' S selected instruction for the first sub-algorithm and the second sub-algorithm, and obtaining a target sub-algorithm according to the selected instruction includes steps S401 to S402:
s401, displaying a first sub-algorithm contained in the first recommendation algorithm and a second sub-algorithm contained in the second recommendation algorithm to a user according to a preset subdivision label;
s402, receiving a selected instruction of the user for the subdivision label, wherein the selected instruction is generated based on clicking operation behavior data;
s403, screening and determining a target sub-algorithm from the first sub-algorithm and the second sub-algorithm according to the clicking operation behavior data.
In this embodiment, in the process of receiving a selected instruction of a user for the first sub-algorithm and the second sub-algorithm and obtaining a target sub-algorithm according to the selected instruction, a first sub-algorithm included in the first recommendation algorithm and a second sub-algorithm included in the second recommendation algorithm are displayed to the user according to a preset subdivision label, that is, by setting a multi-level classification menu in which the first recommendation algorithm and the second recommendation algorithm are configured, and the first sub-algorithm and the second sub-algorithm are configured, and each algorithm in each level of classification menu is displayed with a preset subdivision label, then receiving a selected instruction of the user for the subdivision label, where the selected instruction is generated based on click operation behavior data, and the target sub-algorithm is selected and determined from the first sub-algorithm and the second sub-algorithm according to the click operation behavior data, therefore, a multi-level classification menu is provided for a user to quickly select, the active intention of the user is comprehensively considered, different recommendation algorithms are configured for different users, and the accuracy and diversity of content information personalized pushing are improved.
In one embodiment, as shown in fig. 5, the pushing target content information to the user according to the user portrait tag, the content data tag and the target recommendation algorithm includes the following steps S501-S503:
s501, acquiring recommendation proportions of a plurality of strategies in the target recommendation algorithm;
s502, determining target content information according to the recommended proportion, the user portrait label and the content data label, wherein the target content information comprises a plurality of different amounts of content information;
s503, pushing the corresponding amount of content information to the user.
In this embodiment, in the process of pushing target content information to a user according to the user portrait tag, the content data tag, and the target recommendation algorithm, recommendation proportions of a plurality of strategies in the target recommendation algorithm are obtained, the target recommendation algorithm includes strategies a, b, and c, recommendation proportions corresponding to the strategies a, b, and c are 5,3, and 2, the recommendation proportions are determined based on accuracy of historical statistics of each strategy, then target content information is determined according to the recommendation proportions, the user portrait tag, and the content data tag, the target content information includes a plurality of different amounts of content information, for example, 10 content information is pushed to the user each time, corresponding content information pushed by the strategy a is 5, content information pushed by the strategy b is 3, content information pushed by the strategy c is 2, and then, the corresponding amount of content information is pushed to the user, so that the accuracy and diversity of the personalized pushing of the content information are improved.
In one embodiment, as shown in fig. 6, the obtaining the first recommendation algorithm and the second recommendation algorithm of different policies includes the following steps S601-S604:
s601, acquiring a plurality of algorithms to be selected with different strategies from an algorithm library;
s602, acquiring the historical pushing accuracy of the algorithm to be selected;
s603, sorting the algorithms to be selected according to the historical pushing accuracy;
s604, obtaining the sorted pre-preset number of algorithms to be selected as a first recommendation algorithm and a second recommendation algorithm.
In the embodiment, in the process of acquiring a first recommendation algorithm and a second recommendation algorithm of different strategies, a plurality of algorithms to be selected of different strategies are firstly acquired from an algorithm library, the historical push accuracy of the algorithms to be selected is acquired, and the algorithms to be selected are sorted according to the historical push accuracy; the method comprises the steps of obtaining a preset number of sorted algorithms to be selected as a first recommendation algorithm and a second recommendation algorithm, namely the first recommendation algorithm and the second recommendation algorithm do not designate two recommendation algorithms but designate different recommendation algorithms, the recommendation algorithms comprise a content recommendation algorithm based on a demographic recommendation algorithm, a collaborative filtering recommendation algorithm, a mixed recommendation algorithm of at least two recommendation algorithms, non-recommendation algorithm and the like, sorting is carried out after historical pushing accuracy rates of the different algorithms to be selected are counted through historical data, the recommendation algorithms with higher accuracy can be effectively configured, and users can select the algorithms by themselves, so that accuracy and diversity of content information personalized pushing are improved.
In one embodiment, as shown in fig. 7, after the target content information is pushed to the user according to the user portrait tag, the content data tag and the target recommendation algorithm, the method further includes the following steps S701-S703:
s701, receiving a switching instruction for switching the target recommendation algorithm by a user;
s702, acquiring a first target recommendation algorithm according to the switching instruction;
s703, determining first recommended content information according to the first target recommendation algorithm, the user portrait label and the content data label, and displaying the first recommended content information in a pre-browsing mode.
In this embodiment, after pushing the target content information to the user according to the user portrait tag, the content data tag, and the target recommendation algorithm, the user may change the current recommendation algorithm, that is, receive a switching instruction for switching the target recommendation algorithm by the user, obtain the first target recommendation algorithm according to the switching instruction, the user may reselect a plurality of algorithms to be selected, then regenerate the recommendation algorithm, that is, the first target recommendation algorithm, according to the algorithms to be selected, determine the first recommended content information according to the first target recommendation algorithm, the user portrait tag, and the content data tag, simultaneously display the first recommended content information in a pre-browsing manner, temporarily do not perform global replacement, but previously display the corresponding content information under the algorithm, when the behavior data of the user identifies that the user is interested in the first recommended content information displayed in the pre-browsing manner, and then the global replacement algorithm is the first target recommendation algorithm, so that the accuracy and diversity of the personalized pushing of the content information are improved.
Referring to fig. 8, the present application further provides a content information recommendation apparatus, including:
a behavior data module 101, configured to obtain behavior data of content information browsed by a user;
the data classification module 102 is configured to classify the behavior data and generate a user portrait label and a content data label according to a classification result;
the algorithm obtaining module 103 is configured to obtain a first recommendation algorithm and a second recommendation algorithm of different strategies;
the algorithm screening module 104 is configured to receive a selected instruction of a user for the first recommendation algorithm and the second recommendation algorithm, and determine a target recommendation algorithm according to the selected instruction;
and the information pushing module 105 is used for pushing target content information to the user according to the user portrait label, the content data label and the target recommendation algorithm.
As described above, it is understood that each component of the content information recommendation apparatus proposed in the present application can implement the function of any one of the content information recommendation methods described above.
In one embodiment, after obtaining the first recommendation algorithm and the second recommendation algorithm of different policies, the method further includes:
configuring the first recommendation algorithm according to the user portrait label, wherein the first recommendation algorithm comprises a plurality of first sub-algorithms;
and configuring the second recommendation algorithm according to the content data tag, wherein the second recommendation algorithm comprises a plurality of second sub-algorithms.
In one embodiment, the receiving a selected instruction of the user for the first recommendation algorithm and the second recommendation algorithm, and determining a target recommendation algorithm according to the selected instruction includes:
receiving a selected instruction of a user for the first sub-algorithm and the second sub-algorithm, and obtaining a target sub-algorithm according to the selected instruction;
and determining a target recommendation algorithm according to the target sub-algorithm.
In one embodiment, the receiving a user's selected instruction for the first sub-algorithm and the second sub-algorithm, and obtaining a target sub-algorithm according to the selected instruction includes:
displaying a first sub-algorithm contained in the first recommendation algorithm and a second sub-algorithm contained in the second recommendation algorithm to a user according to a preset subdivision label;
receiving a selected instruction of the user for the subdivision label, wherein the selected instruction is generated based on clicking operation behavior data;
and screening and determining a target sub-algorithm from the first sub-algorithm and the second sub-algorithm according to the clicking operation behavior data.
In one embodiment, the pushing target content information to a user according to the user portrait tag, the content data tag and the target recommendation algorithm includes:
acquiring recommendation proportions of a plurality of strategies in the target recommendation algorithm;
determining target content information according to the recommendation proportion, the user portrait label and the content data label, wherein the target content information comprises a plurality of different amounts of content information;
a corresponding amount of content information is pushed to the user.
In one embodiment, the obtaining the first recommendation algorithm and the second recommendation algorithm of different policies includes:
acquiring a plurality of algorithms to be selected with different strategies from an algorithm library;
acquiring the historical pushing accuracy of the algorithm to be selected;
sorting the algorithms to be selected according to the historical pushing accuracy;
and obtaining the sorted pre-preset number of algorithms to be selected as a first recommendation algorithm and a second recommendation algorithm.
In one embodiment, after pushing the target content information to the user according to the user portrait tag, the content data tag and the target recommendation algorithm, the method further includes:
receiving a switching instruction for switching the target recommendation algorithm by a user;
acquiring a first target recommendation algorithm according to the switching instruction;
and determining first recommended content information according to the first target recommendation algorithm, the user portrait label and the content data label, and displaying the first recommended content information in a pre-browsing mode.
Referring to fig. 9, an embodiment of the present application further provides a computer device, where the computer device may be a mobile terminal, and an internal structure of the computer device may be as shown in fig. 9. The computer equipment comprises a processor, a memory, a network interface, a display device and an input device which are connected through a system bus. Wherein, the network interface of the computer equipment is used for communicating with an external terminal through network connection. The display device of the computer device is used for displaying the offline application. The input device of the computer device is used for receiving the input of the user in offline application. The computer designed processor is used to provide computational and control capabilities. The memory of the computer device includes non-volatile storage media. The non-volatile storage medium stores an operating system, a computer program, and a database. The database of the computer device is used for storing the original data. The computer program is executed by a processor to implement a recommendation method of content information.
The processor executes the method for recommending content information, and the method includes: acquiring behavior data of content information browsed by a user; classifying according to the behavior data, and generating a user portrait label and a content data label according to a classification result; acquiring a first recommendation algorithm and a second recommendation algorithm of different strategies; receiving a selected instruction of a user for a first recommendation algorithm and a second recommendation algorithm, and determining a target recommendation algorithm according to the selected instruction; and pushing target content information to the user according to the user portrait label, the content data label and the target recommendation algorithm.
The computer equipment provides a method for pushing content information to a user by comprehensively considering user initiative intention and user image and content data, firstly, behavior data of content information browsed by the user is obtained, then the behavior data is analyzed, a recommendation algorithm is modified according to the behavior data and recommendation preference of the user, more accurate target content is pushed for the user, specifically, classification is carried out according to the behavior data, a user image label and a content data label are generated according to a classification result, then a first recommendation algorithm and a second recommendation algorithm of different strategies are obtained, then a preset recommendation algorithm (namely the first recommendation algorithm and the second recommendation algorithm) is opened to the user for selection by a platform, the user can select any one or more algorithms from the opened recommendation algorithms as a target recommendation algorithm, and the first recommendation algorithm, the second recommendation algorithm, the target recommendation algorithm and the target recommendation algorithm are received by the user, The recommendation algorithm selected by the user is defined as a target recommendation algorithm, target content information is pushed to the user according to the user portrait label, the content data label and the target recommendation algorithm, different recommendation algorithms are configured for different users, the initiative intention of the user and the user portrait and the content data are comprehensively considered, the Martian effect problem in a recommendation system can be effectively solved, and the accuracy and diversity of personalized pushing of the content information are improved.
An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, the computer program, when executed by the processor, implementing a method for recommending content information, including the steps of: acquiring behavior data of content information browsed by a user; classifying according to the behavior data, and generating a user portrait label and a content data label according to a classification result; acquiring a first recommendation algorithm and a second recommendation algorithm of different strategies; receiving a selected instruction of a user for a first recommendation algorithm and a second recommendation algorithm, and determining a target recommendation algorithm according to the selected instruction; and pushing target content information to the user according to the user portrait label, the content data label and the target recommendation algorithm.
The computer readable storage medium provides a method for pushing content information to a user by comprehensively considering user initiative intention and user image and content data, firstly, behavior data of the content information browsed by the user is obtained, then the behavior data is analyzed, a recommendation algorithm is modified according to the behavior data and recommendation preference of the user, more accurate target content is pushed for the user, specifically, classification is carried out according to the behavior data, a user image label and a content data label are generated according to a classification result, then a first recommendation algorithm and a second recommendation algorithm of different strategies are obtained, a preset recommendation algorithm (namely the first recommendation algorithm and the second recommendation algorithm) is opened to the user for selection by a platform, the user can select any one or more algorithms from the open recommendation algorithm as a target recommendation algorithm, and the first recommendation algorithm of the user is received, The recommendation algorithm selected by the user is defined as a target recommendation algorithm, target content information is pushed to the user according to the user portrait label, the content data label and the target recommendation algorithm, different recommendation algorithms are configured for different users, the initiative intention of the user and the user portrait and the content data are comprehensively considered, the Martian effect problem in a recommendation system can be effectively solved, and the accuracy and diversity of personalized pushing of the content information are improved.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are also included in the scope of the present application.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210285462.7A CN114611004A (en) | 2022-03-22 | 2022-03-22 | Recommendation method and device of content information, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210285462.7A CN114611004A (en) | 2022-03-22 | 2022-03-22 | Recommendation method and device of content information, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114611004A true CN114611004A (en) | 2022-06-10 |
Family
ID=81865394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210285462.7A Pending CN114611004A (en) | 2022-03-22 | 2022-03-22 | Recommendation method and device of content information, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114611004A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117078359A (en) * | 2023-10-16 | 2023-11-17 | 山东大学 | Product recommendation methods, systems, equipment and media based on user group classification |
| CN119003878A (en) * | 2024-08-19 | 2024-11-22 | 北京字跳网络技术有限公司 | Song recommendation method, song recommendation device, electronic equipment, storage medium and program product |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107402961A (en) * | 2017-06-16 | 2017-11-28 | 北京三快在线科技有限公司 | One kind recommends method and device, electronic equipment |
| CN107609198A (en) * | 2017-10-20 | 2018-01-19 | 咪咕互动娱乐有限公司 | One kind recommends method, apparatus and computer-readable recording medium |
| CN109741125A (en) * | 2018-11-27 | 2019-05-10 | 口碑(上海)信息技术有限公司 | Recommend method and device, the storage medium, electronic device of vegetable |
| CN112115363A (en) * | 2020-09-22 | 2020-12-22 | 京东方科技集团股份有限公司 | Recommendation method, computing device and storage medium |
| CN113609375A (en) * | 2021-06-21 | 2021-11-05 | 青岛海尔科技有限公司 | Content recommendation method and device, storage medium and electronic device |
-
2022
- 2022-03-22 CN CN202210285462.7A patent/CN114611004A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107402961A (en) * | 2017-06-16 | 2017-11-28 | 北京三快在线科技有限公司 | One kind recommends method and device, electronic equipment |
| CN107609198A (en) * | 2017-10-20 | 2018-01-19 | 咪咕互动娱乐有限公司 | One kind recommends method, apparatus and computer-readable recording medium |
| CN109741125A (en) * | 2018-11-27 | 2019-05-10 | 口碑(上海)信息技术有限公司 | Recommend method and device, the storage medium, electronic device of vegetable |
| CN112115363A (en) * | 2020-09-22 | 2020-12-22 | 京东方科技集团股份有限公司 | Recommendation method, computing device and storage medium |
| CN113609375A (en) * | 2021-06-21 | 2021-11-05 | 青岛海尔科技有限公司 | Content recommendation method and device, storage medium and electronic device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117078359A (en) * | 2023-10-16 | 2023-11-17 | 山东大学 | Product recommendation methods, systems, equipment and media based on user group classification |
| CN117078359B (en) * | 2023-10-16 | 2024-01-12 | 山东大学 | Product recommendation methods, systems, equipment and media based on user group classification |
| CN119003878A (en) * | 2024-08-19 | 2024-11-22 | 北京字跳网络技术有限公司 | Song recommendation method, song recommendation device, electronic equipment, storage medium and program product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kwon et al. | Accurately or accidentally? Recommendation agent and search experience in over-the-top (OTT) services | |
| CN111105819B (en) | Clipping template recommendation method and device, electronic equipment and storage medium | |
| CN104462292B (en) | Socially collaborative filtering | |
| CN109948023B (en) | Recommended object acquisition method, recommended object acquisition device and storage medium | |
| US9967628B2 (en) | Rating videos based on parental feedback | |
| CN110297848A (en) | Recommended models training method, terminal and storage medium based on federation's study | |
| CN110263244A (en) | Content recommendation method, device, storage medium and computer equipment | |
| JP2024021038A (en) | Inter-domain migration recommendation system based on user interests and preferences | |
| CN105635824A (en) | Personalized channel recommendation method and system | |
| RU2714594C1 (en) | Method and system for determining parameter relevance for content items | |
| CN110209922A (en) | Object recommendation method, apparatus, storage medium and computer equipment | |
| CN109783656B (en) | Recommendation method and system of audio and video data, server and storage medium | |
| Kim et al. | Recommendation system for sharing economy based on multidimensional trust model | |
| CN113157898B (en) | Method and device for recommending candidate questions, computer equipment and storage medium | |
| CN114611004A (en) | Recommendation method and device of content information, computer equipment and storage medium | |
| CN111062527B (en) | A video collection traffic prediction method and device | |
| CN111131901A (en) | Method, apparatus, computer device and storage medium for processing long video data | |
| CN112445970A (en) | Information recommendation method and device, electronic equipment and storage medium | |
| WO2024114034A1 (en) | Content recommendation method and apparatus, device, medium, and program product | |
| CN117972214A (en) | Knowledge-graph-based real-time film recommendation processing method and device | |
| CN117972213A (en) | Knowledge graph-based time-sharing film recommendation processing method and device | |
| CN113806662A (en) | Operation data push control method and device, terminal equipment and storage medium | |
| CN118427430A (en) | Recommendation result ordering method, device, equipment and storage medium thereof | |
| CN114385920B (en) | Cross-domain recommendation method, device, equipment and medium based on multi-task learning | |
| CN119414986A (en) | Multimedia resource display method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20220610 |
|
| WD01 | Invention patent application deemed withdrawn after publication |