[go: up one dir, main page]

CN111008874B - Technology trend prediction method, system and storage medium - Google Patents

Technology trend prediction method, system and storage medium Download PDF

Info

Publication number
CN111008874B
CN111008874B CN201911325447.5A CN201911325447A CN111008874B CN 111008874 B CN111008874 B CN 111008874B CN 201911325447 A CN201911325447 A CN 201911325447A CN 111008874 B CN111008874 B CN 111008874B
Authority
CN
China
Prior art keywords
degree
leading
comment
user
information
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.)
Expired - Fee Related
Application number
CN201911325447.5A
Other languages
Chinese (zh)
Other versions
CN111008874A (en
Inventor
郑范瑛
顾复
吴晓波
顾新建
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.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
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 Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201911325447.5A priority Critical patent/CN111008874B/en
Publication of CN111008874A publication Critical patent/CN111008874A/en
Application granted granted Critical
Publication of CN111008874B publication Critical patent/CN111008874B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION 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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0202—Market predictions or forecasting for commercial activities
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION 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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0282—Rating or review of business operators or products
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

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

Abstract

The invention relates to a technical trend prediction method and a technical trend prediction system, wherein the method comprises the following steps: confirming the technical field to be predicted; selecting a leading enterprise belonging to the technical field and product information of each time stage of the leading enterprise; selecting keywords in the product information to establish a word stock; selecting an online community belonging to the technical field and comment information of all users in the online community; processing is carried out according to the comment information and the word stock, and the association degree of keywords in the word stock and the comment information of the user and the early-late degree of the appearance time are obtained; obtaining a lead degree corresponding to the user according to the association degree and the early-late degree, and determining a lead user from the users according to the lead degree; and acquiring comment information of the leading user, and analyzing the comment information to predict the technical development trend. The invention finds out the leading user through comparing the comments of the user with the historical technical development trend of the product, and the latest dynamics of the leading user are mined and generalized, so that the technical development trend can be accurately and effectively obtained.

Description

Technical trend prediction method, system and storage medium
Technical Field
The present invention relates to the technical field of technical trend prediction, and in particular, to a technical trend prediction method, a system and a storage medium.
Background
The technical trend prediction is to infer and judge the future development trend and level of things through scientific methods and logical reasoning according to the past and present information of the prediction object, and the technical trend is predicted to improve the competitiveness of enterprises and the satisfaction of consumers.
The lead user is the customer who adopts an innovation early in its lifecycle. The strong demand of leading users will become a general market demand in the near future. The leading user has two basic features: the demand of the leading user is leading to the common user; the leading users have strong innovative incentives and higher expected benefits. The dynamic state of the leading user is an important index reflecting the technical trend, and in the prior art, the identification and dynamic acquisition of the leading user are basically carried out by adopting a manual investigation and screening mode, so that the efficiency is low, the obtained result is not representative, and the accurate prediction of the technical trend is difficult to realize.
Disclosure of Invention
Aiming at the problems in the prior art, the invention aims to provide a technical trend prediction method, a technical trend prediction system and a storage medium, which can efficiently and accurately select a leading user and acquire a representative technical trend prediction result according to the dynamic state of the leading user.
In order to achieve the above object, the present invention discloses a technical trend prediction method, comprising:
confirming the technical field to be predicted;
selecting a leading enterprise belonging to the technical field and product information of products at each time stage of the leading enterprise;
selecting keywords in the product information to establish a word stock;
selecting an online community belonging to the technical field and comment information of all users in the online community;
processing according to the comment information and the word stock to obtain the association degree of keywords in the word stock and the comment information of the user and the early and late degree of appearance time;
obtaining a lead degree corresponding to the user according to the association degree and the early-late degree, and determining a lead user from the users according to the lead degree;
and acquiring comment information of the leading user, and analyzing the comment information to predict the technical development trend.
Further, the selecting the leading enterprise belonging to the technical field and the product information of the products of each time stage of the leading enterprise includes:
obtaining product information of products disclosed by the leading enterprises in each time stage based on a crawler method, and extracting attribute features of each product by using a keyword extraction method supervised by a support vector machine to obtain the keywords;
the product information at least comprises one or more of the following: product name, release time, product characteristics, product advantages, and product performance;
the selecting the online communities belonging to the technical field and comment information of all users in the online communities comprises the following steps:
acquiring the comment information based on a crawler method; the comment information includes at least: comment content, and corresponding comment time and user account.
Further, the processing according to the comment information and the word stock to obtain the association degree of the keywords in the word stock and the comment information of the user and the early-late degree of the appearance time includes:
step one: performing one-to-one comparison on the keywords in the word stock and the comment contents by adopting a natural language processing method to obtain a correlation degree matrix reflecting the correlation degree;
step two: and comparing the appearance time of the keywords in the word stock with the comment time of the comment information to obtain an early-late degree matrix reflecting that the appearance time of each comment content is earlier than the appearance time of each keyword.
Further, in the first step, the association degree matrix is:
Figure GDA0002384214320000031
wherein i represents an ith keyword in the word stock, and j represents a jth comment content;
in the second step, the early-late degree matrix is:
Figure GDA0002384214320000032
wherein, the elements in the early-late degree matrix are expressed as: t is t ij Log d, d is the number of days.
Further, the obtaining the lead degree corresponding to the user according to the association degree and the early-late degree, and determining the lead user from the users according to the lead degree includes:
the lead degree calculation formula is as follows:
Figure GDA0002384214320000033
wherein Y represents the lead degree, h ij Elements, t, representing the correlation degree matrix ij Elements representing the early and late degree matrix.
Further, the obtaining the comment information of the leading user, analyzing the comment information to predict the technical development trend, includes:
acquiring the latest comment content published by the leading user;
and obtaining comment keywords in the comment content, and merging and eliminating the redundant keywords.
The invention also discloses a technical trend prediction system, which comprises:
the information acquisition module is used for:
confirming the technical field to be predicted; selecting a leading enterprise belonging to the technical field and product information of products at each time stage of the leading enterprise; selecting keywords in the product information to establish a word stock; selecting an online community belonging to the technical field and comment information of all users in the online community;
the calculation and analysis module is used for:
processing according to the comment information and the word stock to obtain the association degree of keywords in the word stock and the comment information of the user and the early and late degree of appearance time; obtaining a lead degree corresponding to the user according to the association degree and the early-late degree, and determining a lead user from the users according to the lead degree;
and the result processing module is used for acquiring comment information of the leading user and analyzing the comment information to predict the technical development trend.
Further, the information acquisition module is specifically configured to:
obtaining product information of products disclosed by the leading enterprises in each time stage based on a crawler method, and extracting attribute features of each product by using a keyword extraction method supervised by a support vector machine to obtain the keywords;
the product information at least comprises one or more of the following: product name, release time, product characteristics, product advantages, and product performance;
acquiring the comment information based on a crawler method;
the comment information includes at least: comment content, and corresponding comment time and user account.
Further, the computational analysis module comprises:
the first computing unit is used for performing one-to-one comparison on each keyword in the word stock and each comment content by adopting a natural language processing method to obtain a correlation degree matrix reflecting the correlation degree;
and the second calculation unit is used for comparing the appearance time of the keywords in the word stock with the comment time of the comment information to obtain an early-late degree matrix reflecting that the appearance time of each comment content is earlier than the appearance time of each keyword.
Further, the association degree matrix is:
Figure GDA0002384214320000051
wherein i represents an ith keyword in the word stock, and j represents a jth comment content;
the early-late degree matrix is as follows:
Figure GDA0002384214320000052
wherein, the element t in the early and late degree matrix ij Expressed as: t is t ij Log d, d is the number of days.
Further, the computational analysis module further comprises: the comprehensive calculation unit is configured to obtain a lead degree corresponding to the user according to the association degree and the early-late degree, and determine a lead user from the users according to the lead degree, where the comprehensive calculation unit includes:
the lead degree calculation formula is as follows:
Figure GDA0002384214320000053
wherein Y represents the lead degree, h ij Elements, t, representing the correlation degree matrix ij Elements representing the early and late degree matrix.
Further, the result processing module includes:
the first processing unit is used for acquiring comment contents which are newly released by the leading user;
and the second processing unit is used for acquiring comment keywords in the comment content, and merging and eliminating the redundant keywords.
The present invention also discloses a computer readable storage medium having instructions stored therein, which when run on a computer, cause the computer to perform comprising:
the technical trend prediction method as described in each of the above embodiments.
The invention has at least the following beneficial effects:
according to the method for predicting the technical trend, the advanced user is found out through comparison of the comment information of the user and the historical technical development trend of the advanced enterprise product, and the latest dynamics of the advanced user are mined and generalized, so that the technical development trend can be accurately and effectively obtained, and the problems that the technical trend is difficult to predict and user innovation is ignored are solved.
Additional features and advantages of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention, illustrate and explain the invention and are not to be construed as limiting the invention.
In the drawings:
FIG. 1 is a flowchart of a method for predicting technical trends according to an embodiment of the present invention;
fig. 2 is a block diagram of a technical trend prediction system according to an embodiment of the present invention.
Detailed Description
The following describes the detailed implementation of the embodiments of the present invention with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the invention, are not intended to limit the invention.
As shown in fig. 1, the invention discloses a technical trend prediction method, which specifically comprises the following steps:
(1) The technical field to be predicted is confirmed, and the technical field which can be divided by a method/rule can be adopted, for example, the technical field specified by an International Patent Classification (IPC) is used as the field to be selected, and the confirmed technical field with the prediction has a corresponding classification number, so that the management is convenient.
(2) And selecting a leading enterprise belonging to the technical field and product information of products at each time stage of the leading enterprise. Compared with the common enterprises, the products issued by the leading enterprises can reflect the trend of the technical development to a certain extent, and the products can be selected according to the business conditions of the enterprises as the standard by adopting the preferred embodiment of the invention.
(3) And selecting keywords in the product information to establish a word stock. Preferably, the statistical process of the selected product of the lead enterprise needs to use a crawler technology, and product information released in each stage of the product is obtained from homepages and annual reports of the lead enterprise, wherein the product information at least comprises one or more of the following: product name, release time, product characteristics, product advantages and product performance, and extracting attribute characteristics corresponding to each product by using a keyword extraction algorithm supervised by a support vector machine to obtain keywords.
(4) And selecting an online community belonging to the technical field and comment information of all users in the online community. Preferably, the enterprise operation online community and the user self-built online community in the selected field are selected, and the interference effect of the enterprise on user innovation is reduced.
(5) And processing according to the comment information and the word stock to obtain the association degree of the keywords in the word stock and the comment information of the user and the early and late degree of the appearance time. Preferably, a data processing model can be built, the model is trained by a natural language processing method, a word stock reflecting the technical development trend and user comments (comment information) can be imported into the model, and the leading user can be found out through processing modes such as calculation, comparison and the like.
(6) And obtaining the lead degree corresponding to the user according to the association degree and the early-late degree, and determining a lead user from the users according to the lead degree. The association degree reflects the accuracy characteristics of the users on the product or technology prediction, the early and late degrees reflect the foresight and foresight characteristics of the users, the higher the characteristics are, the more obvious the corresponding users are, and the higher the lead degree of the corresponding users is, so that the users with the lead degree higher or more than a certain standard can be selected as the lead users.
(7) And acquiring comment information of the leading user, and analyzing the comment information to predict the technical development trend. In order to ensure the accuracy of prediction, comment information released by the leading user in the latest time can be acquired, and in addition, the latest comment information of the leading user can be acquired in real time to modify/supplement/increase the previous prediction result, so that the prediction result can be advanced from time to time.
In some embodiments of the present invention, the selecting the lead enterprise belonging to the technical field and the product information of the products of each time period of the lead enterprise specifically includes:
obtaining product information of products disclosed by the leading enterprise at each time stage based on a crawler method, and extracting attribute features of each product by using a keyword extraction method supervised by a support vector machine to obtain the keywords
In some embodiments of the present invention, the selecting the online community belonging to the technical field and comment information of all users in the online community specifically includes:
the assessment information is obtained based on a crawler method, wherein the assessment information at least comprises: comment content, corresponding comment time and user account number may further include: including online community name, comment time, endorsement number, forwarding number, etc. Preferably, the method and the system can also use the heat/attention degree reflected in the comment information to divide the weight for calculating the lead degree, and users corresponding to the comment information with the heat/attention degree have higher weight, and the comment information of the users with the higher weight is preferentially adopted under the condition of limited calculation resources or conflict of the comment information.
It should be noted that, the crawler method mentioned in each embodiment of the present invention is a data processing method commonly used in the field of computer networks, and the principles and processes of the present invention are not repeated.
In some embodiments of the present invention, the processing according to the comment information and the word stock, to obtain the association degree and the early-late degree of the appearance time of the keyword in the word stock and the comment information of the user, specifically includes:
step one: performing one-to-one comparison on the keywords in the word stock and the comment contents by adopting a natural language processing method to obtain a correlation degree matrix reflecting the correlation degree;
in the first step, the association degree matrix is:
Figure GDA0002384214320000081
wherein i represents the ith keyword in the word stock, and j represents the jth comment content.
Step two: and comparing the appearance time of the keywords in the word stock with the comment time of the comment information to obtain an early-late degree matrix reflecting that the appearance time of each comment content is earlier than the appearance time of each keyword.
In the second step, the early-late degree matrix is:
Figure GDA0002384214320000091
wherein, the elements in the early-late degree matrix are expressed as: t is t ij Log d, d is the number of days.
Finally, the obtaining the lead degree corresponding to the user according to the association degree and the early-late degree, and determining the lead user from the users according to the lead degree includes:
the lead degree calculation formula is as follows:
Figure GDA0002384214320000092
wherein Y represents the lead degree, h ij Elements, t, representing the correlation degree matrix ij Elements representing the early and late degree matrix.
In some embodiments of the present invention, the obtaining comment information of the lead user, and analyzing the comment information to predict a technical development trend, includes:
acquiring the latest comment content published by the leading user;
and obtaining comment keywords in the comment content, and merging and eliminating the redundant keywords.
As shown in fig. 2, the present invention also discloses a technical trend prediction system, which includes:
the information acquisition module is used for: confirming the technical field to be predicted; selecting a leading enterprise belonging to the technical field and product information of products at each time stage of the leading enterprise; selecting keywords in the product information to establish a word stock; selecting an online community belonging to the technical field and comment information of all users in the online community;
the calculation and analysis module is used for: processing according to the comment information and the word stock to obtain the association degree of keywords in the word stock and the comment information of the user and the early and late degree of appearance time; obtaining a lead degree corresponding to the user according to the association degree and the early-late degree, and determining a lead user from the users according to the lead degree;
and the result processing module is used for acquiring comment information of the leading user and analyzing the comment information to predict the technical development trend.
In some embodiments of the present invention, the information acquisition module is specifically configured to: obtaining product information of products disclosed by the leading enterprises in each time stage based on a crawler method, and extracting attribute features of each product by using a keyword extraction method supervised by a support vector machine to obtain the keywords; the product information at least comprises one or more of the following: product name, release time, product characteristics, product advantages, and product performance; and acquiring the comment information based on a crawler method. The comment information includes at least: comment content, and corresponding comment time and user account.
In some embodiments of the invention, the computational analysis module comprises:
the first computing unit is used for performing one-to-one comparison on each keyword in the word stock and each comment content by adopting a natural language processing method to obtain a correlation degree matrix reflecting the correlation degree;
and the second calculation unit is used for comparing the appearance time of the keywords in the word stock with the comment time of the comment information to obtain an early-late degree matrix reflecting that the appearance time of each comment content is earlier than the appearance time of each keyword.
In some embodiments of the present invention, the association degree matrix is:
Figure GDA0002384214320000101
wherein i represents an ith keyword in the word stock, and j represents a jth comment content;
the early-late degree matrix is as follows:
Figure GDA0002384214320000111
wherein, the element t in the early and late degree matrix ij Expressed as: t is t ij Log d, d is the number of days.
In some embodiments of the invention, the computational analysis module further comprises: the comprehensive calculation unit is configured to obtain a lead degree corresponding to the user according to the association degree and the early-late degree, and determine a lead user from the users according to the lead degree, where the comprehensive calculation unit includes:
the lead degree calculation formula is as follows:
Figure GDA0002384214320000112
wherein Y represents the lead degree, h ij Elements, t, representing the correlation degree matrix ij Elements representing the early and late degree matrix.
In some embodiments of the invention, the result processing module includes:
the first processing unit is used for acquiring comment contents which are newly released by the leading user;
and the second processing unit is used for acquiring comment keywords in the comment content, and merging and eliminating the redundant keywords.
The present invention also discloses a computer readable storage medium having instructions stored therein, which when run on a computer, cause the computer to perform comprising: the technical trend prediction method as described in each of the above embodiments.
The following is one of the preferred embodiments of the present invention, and specifically includes the following steps:
s1: the leading enterprises in the selected fields are obtained, and the calendar products and the attributes thereof are obtained.
The method specifically comprises the following steps:
step S1-1: and counting the leading enterprises in the selected field. The selection of the leading enterprise can be performed according to the business conditions of the enterprise as a standard, for example, the selection can be performed according to the business income and the number of practitioners by referring to the notification of the rule of the division standard of small enterprises in the printing by the department of industry and informatization, the national statistics bureau, the national development and reform committee, and the financial bureau.
Step S1-2: and counting products (product information) of the selected leading enterprises, wherein the products comprise enterprise names, product characteristics, product advantages and product performances, and extracting keywords of the product characteristics, the product advantages and the product performances as word stock. The statistical process of the selected leading enterprise products needs to use a crawler technology, products released at each stage and product attributes thereof, including product names, product characteristics, product advantages and product performances, are obtained from homepages and annual reports of the leading enterprises, and the attribute characteristics of each product are extracted by using a keyword extraction algorithm supervised by a support vector machine.
In step S1, selecting a lead enterprise in the selected field according to the operation condition of the song enterprise in the selected field, crawling the calendar year products and the attributes thereof of the selected enterprise by utilizing a crawler technology, and extracting the attribute characteristics of each product by using a support vector machine method to obtain the idea products and the attributes thereof of the lead enterprise.
Step S2: comments of all users in the online community are obtained.
The method specifically comprises the following steps:
step S2-1: an online community of the selected domain is selected.
In the step, the enterprise operation online community and the user self-built online community in the selected field are selected, and the interference effect of the enterprise on user innovation is reduced.
Step S2-2: and counting comment information of users in the online community, wherein the comment information comprises online community names, user IDs, comment time, comment content, praise numbers and forwarding numbers.
The method comprises the steps that a crawler technology is needed to obtain user comment information in an online community, and crawler results are made into standardized data.
Step S3: and obtaining a user lead measurement model, wherein the model is trained by a natural language processing algorithm, and comprises a technical development trend (word stock) and user comments (comment content) to find out a lead user.
The method comprises the following steps:
step S3-1: and obtaining an association degree matrix of the word library words and the user comments.
In the step, a word2vec method in natural language processing is utilized to compare the ith word in the product attribute word stock with the jth comment of the user one by one to obtain the semantic similarity relationship h of the ith word ij Calculating one by one to obtain an association degree matrix of word library words and user comments:
Figure GDA0002384214320000131
step S3-2: and obtaining a word stock word and appearance time early-late degree matrix of the user comments.
In the step, the ith word in the product attribute word stock is compared with the morning and evening of the occurrence time of the jth comment of the user, if the occurrence time of the jth comment of the user is earlier than the ith word in the product attribute word stock and is ahead for d days, the idea of the user is earlier than the release of the product, the lead degree of the user is higher, and the time lead degree t is the time lead degree ij Calculation is carried out one by one to obtain a word library word and appearance time early-late degree matrix of user comments, =log d:
Figure GDA0002384214320000132
step S3-3: and obtaining the user lead of the online community.
In this step, the model is obtained as:
Figure GDA0002384214320000133
wherein Y represents the user's lead, h ij Representing the association degree of the ith word in the word stock and the jth comment of the user, t ij Representing the early and late degree of occurrence of the ith word and the jth comment of the user in the word stock.
In step S3, the user lead of the online community user can be calculated by using steps S3-1 to S3-3, respectively.
Step S4: and acquiring key points in the latest comments of the leading users as potential development trends of the technology.
The method comprises the following steps:
step S4-1: and acquiring the latest comments of the user.
In the step, the latest comments of the user need to be crawled by a crawler method, wherein the latest comments comprise an online community name, a user ID, comment time, comment content, praise numbers and forwarding numbers.
Step S42: and obtaining keywords in the comments, wherein a supervised keyword extraction method is adopted in the step.
In the step, keywords in the latest comments of the user are acquired by using a Textrank algorithm. And obtaining keywords of the user comments one by one, and then merging and eliminating synonyms of the obtained results to obtain analysis results which are used as technical development trends.
According to the technical trend prediction method provided by the embodiment, the leading user is found out through comparison of the user comments and the historical technical development trend, and the latest comments of the leading user are mined and generalized, so that the technical development trend can be accurately and effectively obtained, and the problems that the technical trend is difficult to predict and user innovation is ignored are solved.
The foregoing details of the optional implementation of the embodiment of the present invention have been described in detail with reference to the accompanying drawings, but the embodiment of the present invention is not limited to the specific details of the foregoing implementation, and various simple modifications may be made to the technical solution of the embodiment of the present invention within the scope of the technical concept of the embodiment of the present invention, and these simple modifications all fall within the protection scope of the embodiment of the present invention.
In addition, the specific features described in the above embodiments may be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, various possible combinations of embodiments of the present invention are not described in detail.
Those skilled in the art will appreciate that all or part of the steps in implementing the methods of the embodiments described above may be implemented by a program stored in a storage medium, including instructions for causing a single-chip microcomputer, chip or processor (processor) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In addition, any combination of various embodiments of the present invention may be performed, so long as the concept of the embodiments of the present invention is not violated, and the disclosure of the embodiments of the present invention should also be considered.

Claims (5)

1.一种技术趋势预测方法，其特征在于，包括：1. A technology trend prediction method, characterized by comprising: 确认待预测的技术领域；Identify the technology areas to be forecasted; 选取属于所述技术领域的领先企业，以及所述领先企业的各时间阶段产品的产品信息；Select leading companies in the technical field and product information of the leading companies' products at different time periods; 选取所述产品信息中的关键词建立词库；Select keywords from the product information to establish a vocabulary; 选取属于所述技术领域的在线社区，以及所述在线社区中的所有用户的评论信息；Selecting an online community belonging to the technical field and comment information of all users in the online community; 根据所述评论信息和所述词库进行处理，获得所述词库中关键词与所述用户的评论信息的关联程度和出现时间的早晚程度；Processing is performed according to the comment information and the word library to obtain the degree of association between the keywords in the word library and the comment information of the user and the early or late appearance time; 根据所述关联程度和所述早晚程度，获得所述用户对应的领先度，并根据所述领先度从所述用户中确定领先用户；According to the association degree and the early and late degree, obtaining the leading degree corresponding to the user, and determining the leading user from the users according to the leading degree; 获取所述领先用户的评论信息，解析所述评论信息以预测技术发展趋势；Obtaining comment information from the leading users, and analyzing the comment information to predict technology development trends; 所述选取属于所述技术领域的领先企业，以及所述领先企业的各时间阶段产品的产品信息，包括：The selected leading enterprises in the technical field and the product information of the leading enterprises' products at different time periods include: 基于爬虫方法获取所述领先企业在各时间阶段公开的产品的产品信息，并利用支持向量机有监督的关键词提取法对各产品的属性特征进行提取以获得所述关键词；Acquire product information of products disclosed by the leading enterprises at various time periods based on a crawler method, and extract attribute features of each product using a supervised keyword extraction method using a support vector machine to obtain the keywords; 所述产品信息至少包括以下的一种或几种：产品名称、发布时间、产品特点、产品优势、产品性能；The product information includes at least one or more of the following: product name, release time, product features, product advantages, and product performance; 所述选取属于所述技术领域的在线社区，以及所述在线社区中的所有用户的评论信息，包括：The selecting of the online community belonging to the technical field and the comment information of all users in the online community includes: 基于爬虫方法获取所述评论信息；所述评论信息至少包括：评论内容，以及对应的评论时间和用户账号；Acquire the comment information based on a crawler method; the comment information at least includes: comment content, and corresponding comment time and user account; 所述根据所述评论信息和所述词库进行处理，获得所述词库中关键词与所述用户的评论信息的关联程度和出现时间的早晚程度，包括：The processing according to the comment information and the word library to obtain the relevance degree and the early or late appearance time of the keywords in the word library and the comment information of the user includes: 步骤一：采用自然语言处理方法，对所述词库中的所述各关键词与所述各评论内容进行一一比对，得到反映关联程度的关联程度矩阵；Step 1: using a natural language processing method, comparing each keyword in the vocabulary with each comment content one by one to obtain a correlation matrix reflecting the correlation degree; 步骤二：对比所述词库中的关键词的所述出现时间与所述评论信息的评论时间，获得反映所述各评论内容出现时间早于所述各关键词出现时间的早晚程度矩阵；Step 2: Compare the appearance time of the keywords in the vocabulary with the comment time of the comment information to obtain a matrix reflecting the degree of appearance time of each comment content being earlier than the appearance time of each keyword; 所述步骤一中，所述关联程度矩阵为：In the step 1, the association degree matrix is:
Figure FDA0004200898860000021
Figure FDA0004200898860000021
 其中，i表示所述词库中的第i个关键词，j表示第j个评论内容；Wherein, i represents the i-th keyword in the vocabulary, and j represents the j-th comment content; 所述步骤二中，所述早晚程度矩阵为：In step 2, the early and late degree matrix is:
Figure FDA0004200898860000022
Figure FDA0004200898860000022
 其中，所述早晚程度矩阵中的元素表示为：tij＝log d，d为天数；The elements in the early and late degree matrix are expressed as: t ij =log d, d is the number of days; 所述根据所述关联程度和所述早晚程度，获得所述用户对应的领先度，并根据所述领先度从所述用户中确定领先用户，包括：The obtaining, according to the association degree and the early and late degree, the leading degree corresponding to the user, and determining the leading user from the users according to the leading degree, comprises: 所述领先度计算公式为：The leading degree calculation formula is:
Figure FDA0004200898860000023
Figure FDA0004200898860000023
 其中，Y表示所述领先度，hij表示所述关联程度矩阵的元素，tij表示早晚程度矩阵的元素。Wherein, Y represents the leading degree, hij represents the element of the association degree matrix, and tij represents the element of the early and late degree matrix. 2.根据权利要求1所述的技术趋势预测方法，其特征在于，所述获取所述领先用户的评论信息，解析所述评论信息以预测技术发展趋势，包括：2. The method for predicting technology trends according to claim 1, wherein obtaining the comment information of the leading user and analyzing the comment information to predict technology development trends comprises: 获取所述领先用户最新发布的评论内容；Obtain the latest comment content published by the leading user; 获取评论内容中的评论关键词，并对冗余的所述关键词进行合并和剔除。Obtain comment keywords in the comment content, and merge and remove redundant keywords. 3.一种技术趋势预测系统，其特征在于，包括：3. A technology trend prediction system, characterized by comprising: 信息获取模块，用于：Information acquisition module, used to: 确认待预测的技术领域；选取属于所述技术领域的领先企业，以及所述领先企业的各时间阶段产品的产品信息；选取所述产品信息中的关键词建立词库；选取属于所述技术领域的在线社区，以及所述在线社区中的所有用户的评论信息；Confirm the technical field to be predicted; select leading enterprises in the technical field and product information of the leading enterprises' products at different time periods; select keywords in the product information to establish a vocabulary; select online communities in the technical field and comment information of all users in the online communities; 计算分析模块，用于：Computational analysis modules for: 根据所述评论信息和所述词库进行处理，获得所述词库中关键词与所述用户的评论信息的关联程度和出现时间的早晚程度；根据所述关联程度和所述早晚程度，获得所述用户对应的领先度，并根据所述领先度从所述用户中确定领先用户；Processing the comment information and the word library to obtain the correlation between the keywords in the word library and the comment information of the user and the early or late degree of appearance; obtaining the leading degree corresponding to the user according to the correlation degree and the early or late degree, and determining the leading user from the users according to the leading degree; 结果处理模块，用于获取所述领先用户的评论信息，解析所述评论信息以预测技术发展趋势；A result processing module, used to obtain the comment information of the leading user and analyze the comment information to predict the technology development trend; 所述信息获取模块，具体用于：The information acquisition module is specifically used for: 基于爬虫方法获取所述领先企业在各时间阶段公开的产品的产品信息，并利用支持向量机有监督的关键词提取法对各产品的属性特征进行提取以获得所述关键词；Acquire product information of products disclosed by the leading enterprises at various time periods based on a crawler method, and extract attribute features of each product using a supervised keyword extraction method using a support vector machine to obtain the keywords; 所述产品信息至少包括以下的一种或几种：产品名称、发布时间、产品特点、产品优势、产品性能；The product information includes at least one or more of the following: product name, release time, product features, product advantages, and product performance; 基于爬虫方法获取所述评论信息；Acquire the comment information based on a crawler method; 所述评论信息至少包括：评论内容，以及对应的评论时间和用户账号；The comment information at least includes: comment content, corresponding comment time and user account; 所述计算分析模块，包括：The computing and analyzing module comprises: 第一计算单元，用于采用自然语言处理方法，对所述词库中的所述各关键词与所述各评论内容进行一一比对，得到反映关联程度的关联程度矩阵；A first calculation unit is used to use a natural language processing method to compare the keywords in the vocabulary with the review content one by one to obtain a correlation degree matrix reflecting the correlation degree; 第二计算单元，用于对比所述词库中的关键词的所述出现时间与所述评论信息的评论时间，获得反映所述各评论内容出现时间早于所述各关键词出现时间的早晚程度矩阵；A second calculation unit is used to compare the appearance time of the keywords in the vocabulary with the comment time of the comment information to obtain a matrix reflecting the appearance time of each comment content earlier than the appearance time of each keyword; 所述关联程度矩阵为：The association degree matrix is:
Figure FDA0004200898860000041
Figure FDA0004200898860000041
 其中，i表示所述词库中的第i个关键词，j表示第j个评论内容；Wherein, i represents the i-th keyword in the vocabulary, and j represents the j-th comment content; 所述早晚程度矩阵为：The early and late degree matrix is:
Figure FDA0004200898860000042
Figure FDA0004200898860000042
 其中，所述早晚程度矩阵中的元素tij表示为：tij＝log d，d为天数；The element t ij in the early and late degree matrix is expressed as: t ij =log d, d is the number of days; 所述计算分析模块还包括：综合计算单元，用于根据所述关联程度和所述早晚程度，获得所述用户对应的领先度，并根据所述领先度从所述用户中确定领先用户，包括：The calculation and analysis module further includes: a comprehensive calculation unit, which is used to obtain the leading degree corresponding to the user according to the association degree and the early and late degree, and determine the leading user from the users according to the leading degree, including: 所述领先度计算公式为：The leading degree calculation formula is:
Figure FDA0004200898860000043
Figure FDA0004200898860000043
 其中，Y表示所述领先度，hij表示所述关联程度矩阵的元素，tij表示早晚程度矩阵的元素。Wherein, Y represents the leading degree, hij represents the element of the association degree matrix, and tij represents the element of the early and late degree matrix. 4.根据权利要求3所述的技术趋势预测系统，其特征在于，所述结果处理模块，包括：4. The technology trend forecasting system according to claim 3, characterized in that the result processing module comprises: 第一处理单元，用于获取所述领先用户最新发布的评论内容；A first processing unit is used to obtain the latest comment content published by the leading user; 第二处理单元，用于获取评论内容中的评论关键词，并对冗余的所述关键词进行合并和剔除。The second processing unit is used to obtain comment keywords in the comment content, and merge and remove redundant keywords. 5.一种计算机可读存储介质，其特征在于，所述存储介质中存储有指令，当其在计算机上运行时，使得计算机执行包括：5. A computer-readable storage medium, characterized in that the storage medium stores instructions, which, when executed on a computer, cause the computer to execute the following: 如权利要求1-2中任一项所述的技术趋势预测方法。A method for predicting technology trends as described in any one of claims 1 to 2.
CN201911325447.5A 2019-12-20 2019-12-20 Technology trend prediction method, system and storage medium Expired - Fee Related CN111008874B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911325447.5A CN111008874B (en) 2019-12-20 2019-12-20 Technology trend prediction method, system and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911325447.5A CN111008874B (en) 2019-12-20 2019-12-20 Technology trend prediction method, system and storage medium

Publications (2)

Publication Number Publication Date
CN111008874A CN111008874A (en) 2020-04-14
CN111008874B true CN111008874B (en) 2023-06-06

Family

ID=70116983

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911325447.5A Expired - Fee Related CN111008874B (en) 2019-12-20 2019-12-20 Technology trend prediction method, system and storage medium

Country Status (1)

Country Link
CN (1) CN111008874B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20180056900A (en) * 2016-11-21 2018-05-30 주식회사 에스씨티 Method for production control system using processing mining

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102004046858B4 (en) * 2004-09-27 2007-03-15 Siemens Ag Method for determining a conductive participant in a network
FR2945882A1 (en) * 2009-05-19 2010-11-26 France Telecom DEVICE AND METHOD FOR PREDICTING COMMENTS ASSOCIATED WITH A PRODUCT.
US20120215597A1 (en) * 2011-02-17 2012-08-23 Bank Of America Corporation System for analyzing social media behavioral influence
US10885131B2 (en) * 2016-09-12 2021-01-05 Ebrahim Bagheri System and method for temporal identification of latent user communities using electronic content
US20180075128A1 (en) * 2016-09-13 2018-03-15 Adobe Systems Incorporated Identifying Key Terms Related to an Entity
TWI612488B (en) * 2016-12-05 2018-01-21 財團法人資訊工業策進會 Computer device and method for predicting market demand of commodities
CN107247813A (en) * 2017-07-26 2017-10-13 北京理工大学 A kind of network struction and evolution method based on weighting technique
CN107943909A (en) * 2017-11-17 2018-04-20 合肥工业大学 User demand trend method for digging and device, storage medium based on comment data
JP6584486B2 (en) * 2017-12-19 2019-10-02 ヤフー株式会社 Prediction device, prediction method, and prediction program
US10360631B1 (en) * 2018-02-14 2019-07-23 Capital One Services, Llc Utilizing artificial intelligence to make a prediction about an entity based on user sentiment and transaction history
CN109783609B (en) * 2018-12-25 2022-01-04 厦门智汇权科技有限公司 Product development auxiliary system and method based on artificial intelligence
CN109871486B (en) * 2019-02-18 2021-04-06 合肥工业大学 Product demand analysis method and system for market-ahead under social media environment

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20180056900A (en) * 2016-11-21 2018-05-30 주식회사 에스씨티 Method for production control system using processing mining

Also Published As

Publication number Publication date
CN111008874A (en) 2020-04-14

Similar Documents

Publication Publication Date Title
Liu et al. Combining enterprise knowledge graph and news sentiment analysis for stock price prediction
US20200192894A1 (en) System and method for using data incident based modeling and prediction
Enouen et al. Sparse interaction additive networks via feature interaction detection and sparse selection
US20070255646A1 (en) Methods and Systems for Multi-Credit Reporting Agency Data Modeling
Behpour et al. Automatic trend detection: Time-biased document clustering
CN112734569A (en) Stock risk prediction method and system based on user portrait and knowledge graph
Fu et al. A sentiment-aware trading volume prediction model for P2P market using LSTM
WO2022271431A1 (en) System and method that rank businesses in environmental, social and governance (esg)
Brunk Structuring business process context information for process monitoring and prediction
Yao et al. Using social media information to predict the credit risk of listed enterprises in the supply chain
Song et al. Incorporating research reports and market sentiment for stock excess return prediction: a case of mainland China
Rui et al. Utilizing internet big data and machine learning for product demand forecasting and analysis of its economic benefits
Gupta et al. Twitter based capital market analysis using cloud statistics
Xu et al. Finarena: A human-agent collaboration framework for financial market analysis and forecasting
CN118095270B (en) A method, device, electronic device and storage medium for constructing a logic analysis diagram
Poornima et al. Prediction of water consumption using machine learning algorithm
CN111008874B (en) Technology trend prediction method, system and storage medium
Ntaliakouras et al. An apache spark methodology for forecasting tourism demand in greece
Wang et al. Stock movement and volatility prediction from tweets, macroeconomic factors and historical prices
Krusinskas et al. THE RESEARCH OF RELIABILITY OF BANKRUPTCY PREDICTION MODELS IN LITHUANIAN COMPANIES.
CN116384750A (en) Method and computing device for generating labeled samples and training risk rating prediction model
CN113849618A (en) Strategy determination method and device based on knowledge graph, electronic equipment and medium
Jose et al. What’s in the Text?—A Text-Based Financial Constraint Index for Indian Manufacturing Firms
Seong et al. Online news-based stock price forecasting considering homogeneity in the industrial sector
CN115222490B (en) A method, apparatus, device, and storage medium for generating export orders based on user profiles.

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
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20230606

CF01 Termination of patent right due to non-payment of annual fee