CN113051482A - Web page search intelligent matching recommendation method based on user feature recognition and behavior analysis - Google Patents
Web page search intelligent matching recommendation method based on user feature recognition and behavior analysis Download PDFInfo
- Publication number
- CN113051482A CN113051482A CN202110442038.4A CN202110442038A CN113051482A CN 113051482 A CN113051482 A CN 113051482A CN 202110442038 A CN202110442038 A CN 202110442038A CN 113051482 A CN113051482 A CN 113051482A
- Authority
- CN
- China
- Prior art keywords
- user
- document
- field
- search
- documents
- 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.)
- Withdrawn
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 50
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012216 screening Methods 0.000 claims abstract description 48
- 238000007405 data analysis Methods 0.000 claims description 36
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 239000003814 drug Substances 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000007792 addition Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR 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
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a web page search intelligent matching recommendation method based on user characteristic identification and behavior analysis, which is characterized by comprising the steps of obtaining search words input by a user in a document web page search bar, counting each meaning in the search words input by the user, screening each meaning search word input by the user in each field of each related document, obtaining each attribute data of each meaning search word input by the user in each field of each related document, calculating the weight value of each meaning search word input by the user in each field of each related document, simultaneously counting each document data in each field of a history browsing record of the user, analyzing the preference coefficient of the user to each field document, screening the field document with the highest preference coefficient, calculating the recommendation prediction coefficient of each meaning search word input by the user in each related document in the favorite field document, and comparing and screening each meaning search word input by the user in the favorite field of each related document before.
Description
Technical Field
The invention relates to the technical field of webpage search recommendation, in particular to an intelligent webpage search matching recommendation method based on user characteristic identification and behavior analysis.
Background
With the development and popularization of internet technology, document web page search has almost become a necessary way for modern people to acquire information and knowledge in work and life. In the traditional document web page search, the server searches according to the search words input by the user and feeds back the information of each document matched with the search words to the user, however, the search words may contain a plurality of meanings, the server can not recommend according to the meaning of each word by classification, thereby increasing the load of the server in the searching process, leading to that the document information required by the user is deeply buried in a plurality of documents and is difficult to be quickly obtained, increasing the time for the user to obtain the required document information, reducing the efficiency for the user to obtain the required document, simultaneously, the traditional document web searching can not accurately recommend the document in the favorite field of the user, thereby causing the recommended documents not to meet the requirements of the users, reducing the experience and satisfaction of the users in searching the documents and the web pages, in order to solve the problems, an intelligent webpage search matching recommendation method based on user feature recognition and behavior analysis is designed.
Disclosure of Invention
The invention aims to provide an intelligent matching recommendation method for web page search based on user characteristic identification and behavior analysis, which comprises the steps of obtaining search words input by a user in a document web page search bar, counting each meaning in the search words input by the user, screening each meaning search word input by the user in each field of each related document, obtaining each attribute data of each meaning search word input by the user in each field of each related document, calculating the weight value of each meaning search word input by the user in each field of each related document, simultaneously obtaining the history browsing record of the user in a document web page, counting each document data in each field in the history browsing record of the user, analyzing the preference coefficient of the user to each field of documents, screening the field documents with the highest preference coefficient of the user, and comprehensively calculating the recommendation prediction coefficient of each meaning search word input by the user in each related document in the preference field of each field of the document, the method and the device perform document recommendation by comparing and screening various semantic search words input by the user with various related documents ranked at the top in the favorite domain documents, and solve the problems in the background technology.
The purpose of the invention can be realized by the following technical scheme:
the intelligent matching recommendation method for webpage search based on user feature recognition and behavior analysis comprises the following steps:
s1, acquiring search words input by a user in a document webpage search bar, and counting the meaning of each word in the search words input by the user;
s2, screening relevant documents of each meaning search word input by the user in each field, and acquiring attribute data of the relevant documents of each meaning search word input by the user in each field;
s3, calculating the weight value of each relevant document in each field of each meaning search word input by the user;
s4, acquiring the historical browsing records of the user in the document webpage, and counting the document data of each field in the historical browsing records of the user;
s5, analyzing the user' S preference coefficient for each domain document, and screening the domain document with the highest user preference coefficient;
s6, comprehensively calculating the recommendation prediction coefficients of each related document in the favorite domain documents of each semantic search word input by the user;
s7, comparing and screening various semantic search words input by the user and recommending documents in various related documents ranked at the top in the favorite field documents;
the webpage searching intelligent matching recommendation method based on the user characteristic identification and the behavior analysis uses a webpage searching intelligent matching recommendation system based on the user characteristic identification and the behavior analysis, and comprises a searching word acquisition module, a word meaning statistic module, a document screening module, an attribute data acquisition module, an attribute data analysis module, a history record acquisition module, a history data statistic module, a history data analysis module, a cloud storage database, an analysis server and a cloud recommendation platform;
the semantic meaning statistic module is respectively connected with the search word acquisition module and the document screening module, the attribute data acquisition module is respectively connected with the document screening module and the attribute data analysis module, the attribute data analysis module is respectively connected with the cloud storage database and the analysis server, the historical data statistic module is respectively connected with the historical record acquisition module and the historical data analysis module, the historical data analysis module is respectively connected with the cloud storage database and the analysis server, and the analysis server is respectively connected with the cloud storage database and the cloud recommendation platform;
the search word acquisition module is used for acquiring search words input by a user in a document webpage search bar and sending the search words input by the user in the document webpage search bar to the lexical meaning statistics module;
the word meaning counting module is used for receiving the search words input by the user in the document webpage search bar and sent by the search word acquisition module, inquiring and acquiring the meaning of each word in the search words input by the user in the document webpage search bar, counting each word meaning in the search words input by the user, and forming a word meaning set A (a) in the search words input by the user1,a2,...,ai,...,an),aiThe meaning of the ith word in the search words input by the user is represented, and each meaning set in the search words input by the user is sent to a document screening module;
the document screening module is used for receiving each word meaning set in the search words input by the user and sent by the word meaning statistics module, screening each relevant document of each word meaning search word input by the user in each field, and forming each relevant document set W of each word meaning search word input by the user in each fieldij(wij1,wij2,...,wijk,...,wijf),wijkThe method comprises the steps that the ith word meaning search word input by a user is represented as the kth related document in the jth field, j is 1,2, and m, and each word meaning search word input by the user in each field is sent to an attribute data acquisition module;
the attribute data acquisition module is used for receiving each relevant document set of each semantic search word input by a user in each field, which is sent by the document screening module, respectively acquiring the browsing amount, the collection amount and the downloading amount of each semantic search word input by the user in each relevant document in each field, counting each attribute data of each semantic search word input by the user in each relevant document in each field, and sending each attribute data of each semantic search word input by the user in each relevant document in each field to the attribute data analysis module;
the attribute data analysis module is used for receiving each attribute data of each relevant document of each semantic search word input by the user in each field, which is sent by the attribute data acquisition module, extracting a weight influence proportion coefficient corresponding to a document browsing amount, a collection amount and a downloading amount stored in the cloud storage database, calculating a weight value of each semantic search word input by the user in each relevant document in each field, and sending each semantic search word input by the user to the weight value analysis server of each relevant document in each field;
the history record acquisition module is used for acquiring the history browsing record of the user in the document webpage and sending the history browsing record of the user in the document webpage to the history data statistics module;
the historical data counting module is used for receiving the historical browsing records of the user in the document webpage, which are sent by the historical browsing record obtaining module, extracting the document browsing amount, the document collection amount, the document downloading amount and the document browsing time of each field in the historical browsing records of the user, counting the document data of each field in the historical browsing records of the user, and sending the document data of each field in the historical browsing records of the user to the historical data analysis module;
the historical data analysis module is used for receiving the document data of each field in the historical browsing record of the user sent by the historical data statistics module, extracting the preference influence proportion coefficient corresponding to the document browsing amount, the document collection amount, the document downloading amount and the document browsing time stored in the cloud storage database, calculating the preference coefficient of the user to the documents of each field, counting the preference coefficient of the user to the documents of each field, and sending the preference coefficient of the user to the documents of each field to the analysis server;
the cloud storage database is used for storing weight influence proportion coefficients corresponding to the document browsing amount, the document collection amount and the document downloading amount, and the weight influence proportion coefficients are respectively recorded as lambdar′,λr″,λr″′Simultaneously storing the corresponding preference degree influence proportion coefficient of each document data and recording as mup,p=p1,p2,p3,p4;
The analysis server is used for receiving the weighted values of all relevant documents of all the semantic search words input by the user in all the fields, which are sent by the attribute data analysis module, receiving the love coefficient of the user to the documents in all the fields, which is sent by the historical data analysis module, screening the documents in the fields with the highest love coefficient of the user, calculating the recommendation prediction coefficient of all the semantic search words input by the user in all the relevant documents in the love field documents, counting the recommendation prediction coefficient of all the semantic search words input by the user in all the relevant documents in the love field documents, and sending the recommendation prediction coefficient of all the semantic search words input by the user in all the relevant documents in the love field documents to the cloud recommendation platform;
the cloud recommendation platform is used for receiving recommendation prediction coefficients of all relevant documents in the favorite field documents of all the semantic search words input by the user and sent by the analysis server, arranging the recommendation prediction coefficients of all the relevant documents in the favorite field documents of all the semantic search words input by the user, sequentially arranging the recommendation prediction coefficients from large to small, screening all relevant documents with top ranking in the favorite field documents of all the semantic search words input by the user, and sequentially recommending the documents.
Further, the fields in the document screening module respectively comprise agriculture, industry, medicine, military affairs, building, politics, science and technology, astronomy and geography.
Further, the attribute data acquisition module comprises attribute data sets of related documents in various fields, wherein the attribute data sets form various meaning search words input by the user And r attribute data of the kth related document in the jth field, wherein the ith term search term input by the user is represented as r ', r ', r ' are respectively represented as browsing amount, collection amount and downloading amount of the related document.
Further, each meaning search word input by the user is in each fieldThe weighted value of each related document is calculated according to the formulaξijkExpressing the weight value, lambda, of the kth related document in the jth field of the ith semantic search word input by the userr′,λr″,λr″′Respectively expressed as the weight influence proportional coefficients corresponding to the document browsing amount, the collection amount and the downloading amount,representing the browsing amount of the kth related document in the jth field for the ith semantic search word input by the user,representing the collection of the kth relevant document in the jth field for the ith semantic search term input by the user,and expressing the download amount of the kth related document in the jth field of the ith semantic search word input by the user.
Furthermore, the history record acquisition module accesses the user account, checks the history browsing document record of the user account in the document webpage in the last month, and acquires the history browsing record of the user in the document webpage.
Further, the historical data statistical module comprises each document data set XP (x) forming each field in the historical browsing records of the user1p,x2p,...,xjp,...,xmp),xjp is the p-th document data of the j-th field in the history browsing record of the user, and p is p1,p2,p3,p4,p1,p2,p3,p4Respectively expressed as document browsing amount, document collection amount, document downloading amount and document browsing time.
Further, the user's preference coefficient calculation formula for each domain document isψjExpressed as the user's preference coefficient, μ, for the jth domain documentpExpressing the likeness influence proportional coefficient, x, corresponding to the pth document datajp is expressed as the p-th document data of the j-th field in the history browsing record of the user.
Further, the analysis server arranges the preference coefficients of the user for the documents in each field according to the sequence of the preference coefficients from large to small, screens the document in the field with the highest preference coefficient of the user, and records the document as j ', wherein j' belongs to m.
Furthermore, the calculation formula of the recommendation prediction coefficient of each related document in the favorite domain documents of each semantic search word input by the user is The recommendation prediction coefficient, xi, of the kth related document in the favorite domain document of the ith semantic search word input by the userij′kThe weight value of the kth related document in the favorite field of the ith semantic search word input by the user is expressed, e is a natural number and is equal to 2.718, psij′And representing the preference coefficient of the user favorite domain document.
Has the advantages that:
(1) the invention provides a web page search intelligent matching recommendation method based on user characteristic identification and behavior analysis, which comprises the steps of obtaining search words input by a user in a document web page search bar, counting each meaning in the search words input by the user, screening each meaning search word input by the user in each field of each related document, obtaining each attribute data of each meaning search word input by the user in each field of each related document, calculating the weight value of each meaning search word input by the user in each field of each related document, providing reliable reference data for later-stage calculation of recommendation prediction coefficients of each meaning search word input by the user in each related document in favorite fields, simultaneously obtaining historical browsing records of the user in a document web page, counting each document data in each field in the historical browsing records of the user, analyzing the favorite coefficients of the user to each field of documents, and screening the domain documents with the highest user preference coefficient, so that the recommended documents meet the requirements of the user, and the document webpage searching experience and satisfaction of the user are improved.
(2) According to the method and the device, the recommendation estimation coefficients of all relevant documents in the favorite field documents of all the semantic search words input by the user are comprehensively calculated, and the documents are recommended by comparing and screening all the semantic search words input by the user with all relevant documents in the favorite field documents which are ranked at the front, so that the load of a server in the searching process is reduced, the server performs classified recommendation according to all the semantic search words, the document information required by the user can be quickly acquired, the time for the user to acquire the required document information is reduced, and the efficiency for the user to acquire the required documents is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of the process steps of the present invention;
fig. 2 is a schematic view of a module connection structure according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the method for recommending web search intelligent matching based on user feature recognition and behavior analysis includes the following steps:
s1, acquiring search words input by a user in a document webpage search bar, and counting the meaning of each word in the search words input by the user;
s2, screening relevant documents of each meaning search word input by the user in each field, and acquiring attribute data of the relevant documents of each meaning search word input by the user in each field;
s3, calculating the weight value of each relevant document in each field of each meaning search word input by the user;
s4, acquiring the historical browsing records of the user in the document webpage, and counting the document data of each field in the historical browsing records of the user;
s5, analyzing the user' S preference coefficient for each domain document, and screening the domain document with the highest user preference coefficient;
s6, comprehensively calculating the recommendation prediction coefficients of each related document in the favorite domain documents of each semantic search word input by the user;
and S7, comparing and screening the semantic search words input by the user and recommending the related documents ranked at the top in the favorite domain documents.
Referring to fig. 2, the web search intelligent matching recommendation method based on user feature recognition and behavior analysis uses a web search intelligent matching recommendation system based on user feature recognition and behavior analysis, which includes a search term acquisition module, a word meaning statistics module, a document screening module, an attribute data acquisition module, an attribute data analysis module, a history record acquisition module, a history data statistics module, a history data analysis module, a cloud storage database, an analysis server, and a cloud recommendation platform.
The semantic meaning statistic module is respectively connected with the search word acquisition module and the document screening module, the attribute data acquisition module is respectively connected with the document screening module and the attribute data analysis module, the attribute data analysis module is respectively connected with the cloud storage database and the analysis server, the historical data statistic module is respectively connected with the historical record acquisition module and the historical data analysis module, the historical data analysis module is respectively connected with the cloud storage database and the analysis server, and the analysis server is respectively connected with the cloud storage database and the cloud recommendation platform.
The search word acquisition module is used for acquiring search words input by a user in the document webpage search bar and sending the search words input by the user in the document webpage search bar to the word meaning statistics module.
The word meaning counting module is used for receiving the search words input by the user in the document webpage search bar and sent by the search word acquisition module, inquiring and acquiring the meaning of each word in the search words input by the user in the document webpage search bar, counting each word meaning in the search words input by the user, and forming a word meaning set A (a) in the search words input by the user1,a2,...,ai,...,an),aiThe meaning is represented as the ith meaning in the search words input by the user, and each meaning set in the search words input by the user is sent to the document screening module.
The document screening module is used for receiving each word meaning set in the search words input by the user and sent by the word meaning statistics module, screening each relevant document of each word meaning search word input by the user in each field, and forming each relevant document set W of each word meaning search word input by the user in each fieldij(wij1,wij2,...,wijk,...,wijf),wijkAnd sending each set of relevant documents in each field of each lexeme search word input by the user to the attribute data acquisition module.
The fields in the document screening module respectively comprise agriculture, industry, medicine, military, construction, politics, science and technology, astronomy and geography.
The attribute data acquisition module is used for receiving the related document sets of the semantic search words input by the user in the fields sent by the document screening module, and respectively acquiring the browsing amount of the related documents of the semantic search words input by the user in the fields,Collecting and downloading amount, counting attribute data of each related document in each field of each semantic search word input by user, and forming attribute data set of each related document in each field of each semantic search word input by user And r attribute data of the kth related document in the jth field, wherein r is r ', r ', r ' is respectively expressed as browsing amount, collection amount and downloading amount of the related document, and each attribute data of each related document in each field of each semantic search word input by the user is sent to the attribute data analysis module.
The attribute data analysis module is used for receiving the attribute data of each relevant document of each meaning search word input by the user in each field sent by the attribute data acquisition module, extracting the weight influence proportion coefficient corresponding to the browsing amount, the collection amount and the downloading amount of the document stored in the cloud storage database, and calculating the weight value of each relevant document of each meaning search word input by the user in each fieldξijkExpressing the weight value, lambda, of the kth related document in the jth field of the ith semantic search word input by the userr′,λr″,λr″′Respectively expressed as the weight influence proportional coefficients corresponding to the document browsing amount, the collection amount and the downloading amount,representing the browsing amount of the kth related document in the jth field for the ith semantic search word input by the user,indicating the kth piece of the ith semantic search word input by the user in the jth fieldThe collection of the relevant documents is made,and the method comprises the steps of expressing the download amount of the ith semantic search word input by a user in the kth related document in the jth field, sending the weighted value of each semantic search word input by the user in each field to an analysis server, and providing reliable reference data for calculating the recommendation prediction coefficient of each semantic search word input by the user in each related document in the favorite field.
The history record acquisition module is used for acquiring the history browsing record of the user in the document webpage, viewing the history browsing record of the user account in the document webpage within the next month by accessing the user account, acquiring the history browsing record of the user in the document webpage, and sending the history browsing record of the user in the document webpage to the history data statistics module.
The historical data counting module is used for receiving the historical browsing records of the user in the document webpage sent by the historical record acquisition module, extracting the document browsing amount, the document collection amount, the document downloading amount and the document browsing time of each field in the historical browsing records of the user, counting the document data of each field in the historical browsing records of the user, and forming each document data set XP (x) of each field in the historical browsing records of the user1p,x2p,...,xjp,...,xmp),xjp is the p-th document data of the j-th field in the history browsing record of the user, and p is p1,p2,p3,p4,p1,p2,p3,p4And the data are respectively expressed as document browsing amount, document collection amount, document downloading amount and document browsing time, and the document data of each field in the historical browsing record of the user are sent to the historical data analysis module.
The historical data analysis module is used for receiving the document data of each field in the historical browsing record of the user sent by the historical data statistical module, and extracting the corresponding document browsing amount, document collection amount, document downloading amount and document browsing time stored in the cloud storage databaseThe likeness influence proportion coefficient is calculated, and the likeness coefficient of the user to the documents in each field is calculatedψjExpressed as the user's preference coefficient, μ, for the jth domain documentpExpressing the likeness influence proportional coefficient, x, corresponding to the pth document datajAnd p is expressed as the p-th document data of the jth field in the historical browsing record of the user, the preference coefficient of the user to the documents of each field is counted, and the preference coefficient of the user to the documents of each field is sent to the analysis server.
The cloud storage database is used for storing weight influence proportion coefficients corresponding to the document browsing amount, the document collection amount and the document downloading amount, and the weight influence proportion coefficients are respectively recorded as lambdar′,λr″,λr″′Simultaneously storing the corresponding preference degree influence proportion coefficient of each document data and recording as mup,p=p1,p2,p3,p4。
The analysis server is used for receiving the weighted values of all the relevant documents of all the semantic search words input by the user in all the fields sent by the attribute data analysis module, receiving the preference coefficients of the user to the documents in all the fields sent by the historical data analysis module, arranging the preference coefficients of the user to the documents in all the fields according to the sequence of the preference coefficients from large to small, screening the document in the field with the highest preference coefficient of the user and marking the document as j ', wherein j' belongs to m, so that the recommended document meets the requirements of the user, the document webpage search experience and satisfaction of the user are improved, and the recommendation prediction coefficient of all the relevant documents of all the semantic search words input by the user in the document in the favorite field is calculated The recommendation prediction coefficient, xi, of the kth related document in the favorite domain document of the ith semantic search word input by the userij′kExpressed as user inputThe weighted value of the kth related document of the i meaning search words in the favorite fields, e is expressed as a natural number and is equal to 2.718, psij′The recommendation prediction coefficients of all the related documents in the favorite field documents of all the semantic search words input by the user are counted, and the recommendation prediction coefficients of all the related documents in the favorite field documents of all the semantic search words input by the user are sent to the cloud recommendation platform.
The cloud recommendation platform is used for receiving recommendation prediction coefficients of all relevant documents in favorite field documents of all meaning search words input by a user and sent by an analysis server, arranging the recommendation prediction coefficients of all relevant documents in the favorite field documents of all meaning search words input by the user, sequentially arranging the recommendation prediction coefficients according to the descending order of the recommendation prediction coefficients, screening all relevant documents with the meaning search words input by the user, which are ranked at the front in the favorite field documents, and sequentially recommending the documents, so that the load of the server in the search process is reduced, the server performs classification recommendation according to all the meaning search words, document information required by the user can be rapidly obtained, the time for the user to obtain required document information is reduced, and the efficiency for the user to obtain the required documents is improved.
The foregoing is merely exemplary and illustrative of the principles of the present invention and various modifications, additions and substitutions of the specific embodiments described herein may be made by those skilled in the art without departing from the principles of the present invention or exceeding the scope of the claims set forth herein.
Claims (9)
1. The intelligent webpage search matching recommendation method based on user feature recognition and behavior analysis is characterized by comprising the following steps of: the method comprises the following steps:
s1, acquiring search words input by a user in a document webpage search bar, and counting the meaning of each word in the search words input by the user;
s2, screening relevant documents of each meaning search word input by the user in each field, and acquiring attribute data of the relevant documents of each meaning search word input by the user in each field;
s3, calculating the weight value of each relevant document in each field of each meaning search word input by the user;
s4, acquiring the historical browsing records of the user in the document webpage, and counting the document data of each field in the historical browsing records of the user;
s5, analyzing the user' S preference coefficient for each domain document, and screening the domain document with the highest user preference coefficient;
s6, comprehensively calculating the recommendation prediction coefficients of each related document in the favorite domain documents of each semantic search word input by the user;
s7, comparing and screening various semantic search words input by the user and recommending documents in various related documents ranked at the top in the favorite field documents;
the webpage searching intelligent matching recommendation method based on the user characteristic identification and the behavior analysis uses a webpage searching intelligent matching recommendation system based on the user characteristic identification and the behavior analysis, and comprises a searching word acquisition module, a word meaning statistic module, a document screening module, an attribute data acquisition module, an attribute data analysis module, a history record acquisition module, a history data statistic module, a history data analysis module, a cloud storage database, an analysis server and a cloud recommendation platform;
the semantic meaning statistic module is respectively connected with the search word acquisition module and the document screening module, the attribute data acquisition module is respectively connected with the document screening module and the attribute data analysis module, the attribute data analysis module is respectively connected with the cloud storage database and the analysis server, the historical data statistic module is respectively connected with the historical record acquisition module and the historical data analysis module, the historical data analysis module is respectively connected with the cloud storage database and the analysis server, and the analysis server is respectively connected with the cloud storage database and the cloud recommendation platform;
the search word acquisition module is used for acquiring search words input by a user in a document webpage search bar and sending the search words input by the user in the document webpage search bar to the lexical meaning statistics module;
the word meaning statistic moduleThe system is used for receiving the search words input by the user in the document webpage search bar sent by the search word acquisition module, inquiring and acquiring the meanings of all the words in the search words input by the user in the document webpage search bar, counting the meanings of all the words in the search words input by the user, and forming an idea set A (a) in the search words input by the user1,a2,...,ai,...,an),aiThe meaning of the ith word in the search words input by the user is represented, and each meaning set in the search words input by the user is sent to a document screening module;
the document screening module is used for receiving each word meaning set in the search words input by the user and sent by the word meaning statistics module, screening each relevant document of each word meaning search word input by the user in each field, and forming each relevant document set W of each word meaning search word input by the user in each fieldij(wij1,wij2,...,wijk,...,wijf),wijkThe method comprises the steps that the ith word meaning search word input by a user is represented as the kth related document in the jth field, j is 1,2, and m, and each word meaning search word input by the user in each field is sent to an attribute data acquisition module;
the attribute data acquisition module is used for receiving each relevant document set of each semantic search word input by a user in each field, which is sent by the document screening module, respectively acquiring the browsing amount, the collection amount and the downloading amount of each semantic search word input by the user in each relevant document in each field, counting each attribute data of each semantic search word input by the user in each relevant document in each field, and sending each attribute data of each semantic search word input by the user in each relevant document in each field to the attribute data analysis module;
the attribute data analysis module is used for receiving each attribute data of each relevant document of each semantic search word input by the user in each field, which is sent by the attribute data acquisition module, extracting a weight influence proportion coefficient corresponding to a document browsing amount, a collection amount and a downloading amount stored in the cloud storage database, calculating a weight value of each semantic search word input by the user in each relevant document in each field, and sending each semantic search word input by the user to the weight value analysis server of each relevant document in each field;
the history record acquisition module is used for acquiring the history browsing record of the user in the document webpage and sending the history browsing record of the user in the document webpage to the history data statistics module;
the historical data counting module is used for receiving the historical browsing records of the user in the document webpage, which are sent by the historical browsing record obtaining module, extracting the document browsing amount, the document collection amount, the document downloading amount and the document browsing time of each field in the historical browsing records of the user, counting the document data of each field in the historical browsing records of the user, and sending the document data of each field in the historical browsing records of the user to the historical data analysis module;
the historical data analysis module is used for receiving the document data of each field in the historical browsing record of the user sent by the historical data statistics module, extracting the preference influence proportion coefficient corresponding to the document browsing amount, the document collection amount, the document downloading amount and the document browsing time stored in the cloud storage database, calculating the preference coefficient of the user to the documents of each field, counting the preference coefficient of the user to the documents of each field, and sending the preference coefficient of the user to the documents of each field to the analysis server;
the cloud storage database is used for storing weight influence proportion coefficients corresponding to the document browsing amount, the document collection amount and the document downloading amount, and the weight influence proportion coefficients are respectively recorded as lambdar′,λr″,λr″′Simultaneously storing the corresponding preference degree influence proportion coefficient of each document data and recording as mup,p=p1,p2,p3,p4;
The analysis server is used for receiving the weighted values of all relevant documents of all the semantic search words input by the user in all the fields, which are sent by the attribute data analysis module, receiving the love coefficient of the user to the documents in all the fields, which is sent by the historical data analysis module, screening the documents in the fields with the highest love coefficient of the user, calculating the recommendation prediction coefficient of all the semantic search words input by the user in all the relevant documents in the love field documents, counting the recommendation prediction coefficient of all the semantic search words input by the user in all the relevant documents in the love field documents, and sending the recommendation prediction coefficient of all the semantic search words input by the user in all the relevant documents in the love field documents to the cloud recommendation platform;
the cloud recommendation platform is used for receiving recommendation prediction coefficients of all relevant documents in the favorite field documents of all the semantic search words input by the user and sent by the analysis server, arranging the recommendation prediction coefficients of all the relevant documents in the favorite field documents of all the semantic search words input by the user, sequentially arranging the recommendation prediction coefficients from large to small, screening all relevant documents with top ranking in the favorite field documents of all the semantic search words input by the user, and sequentially recommending the documents.
2. The web search intelligent matching recommendation method based on user feature recognition and behavior analysis according to claim 1, characterized in that: the fields in the document screening module respectively comprise agriculture, industry, medicine, military, construction, politics, science and technology, astronomy and geography.
3. The web search intelligent matching recommendation method based on user feature recognition and behavior analysis according to claim 1, characterized in that: the attribute data acquisition module comprises attribute data sets of related documents in various fields of various meaning search words input by a user And r attribute data of the kth related document in the jth field, wherein the ith term search term input by the user is represented as r ', r ', r ' are respectively represented as browsing amount, collection amount and downloading amount of the related document.
4. The method according to claim 3The intelligent matching recommendation method for webpage search based on user feature recognition and behavior analysis is characterized by comprising the following steps of: the weighted value calculation formula of each related document in each field of each meaning search word input by the user isξijkExpressing the weight value, lambda, of the kth related document in the jth field of the ith semantic search word input by the userr′,λr″,λr″′Respectively expressed as the weight influence proportional coefficients corresponding to the document browsing amount, the collection amount and the downloading amount,representing the browsing amount of the kth related document in the jth field for the ith semantic search word input by the user,representing the collection of the kth relevant document in the jth field for the ith semantic search term input by the user,and expressing the download amount of the kth related document in the jth field of the ith semantic search word input by the user.
5. The web search intelligent matching recommendation method based on user feature recognition and behavior analysis according to claim 1, characterized in that: the history record acquisition module accesses the user account, checks the history browsing document record of the user account in the document webpage in the last month and acquires the history browsing record of the user in the document webpage.
6. The web search intelligent matching recommendation method based on user feature recognition and behavior analysis according to claim 1, characterized in that: the historical data statistical module comprises various collars in the historical browsing records of the userEach document data set XP (x) of a domain1p,x2p,...,xjp,...,xmp),xjp is the p-th document data of the j-th field in the history browsing record of the user, and p is p1,p2,p3,p4,p1,p2,p3,p4Respectively expressed as document browsing amount, document collection amount, document downloading amount and document browsing time.
7. The web search intelligent matching recommendation method based on user feature recognition and behavior analysis of claim 6, characterized in that: the user's likeness coefficient calculation formula for each field document isψjExpressed as the user's preference coefficient, μ, for the jth domain documentpExpressing the likeness influence proportional coefficient, x, corresponding to the pth document datajp is expressed as the p-th document data of the j-th field in the history browsing record of the user.
8. The web search intelligent matching recommendation method based on user feature recognition and behavior analysis of claim 7, wherein: the analysis server arranges the preference coefficients of the user to the documents in each field according to the sequence of the preference coefficients from large to small, screens the document in the field with the highest preference coefficient of the user, and records the document as j ', wherein j' belongs to m.
9. The web search intelligent matching recommendation method based on user feature recognition and behavior analysis of claim 8, wherein: the recommendation prediction coefficient calculation formula of each related document in the favorite domain documents of each meaning search word input by the user is The recommendation prediction coefficient, xi, of the kth related document in the favorite domain document of the ith semantic search word input by the userij′kThe weight value of the kth related document in the favorite field of the ith semantic search word input by the user is expressed, e is a natural number and is equal to 2.718, psij′And representing the preference coefficient of the user favorite domain document.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110442038.4A CN113051482A (en) | 2021-04-23 | 2021-04-23 | Web page search intelligent matching recommendation method based on user feature recognition and behavior analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110442038.4A CN113051482A (en) | 2021-04-23 | 2021-04-23 | Web page search intelligent matching recommendation method based on user feature recognition and behavior analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113051482A true CN113051482A (en) | 2021-06-29 |
Family
ID=76520053
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110442038.4A Withdrawn CN113051482A (en) | 2021-04-23 | 2021-04-23 | Web page search intelligent matching recommendation method based on user feature recognition and behavior analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113051482A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114861019A (en) * | 2022-05-03 | 2022-08-05 | 北京博智瑞成科技有限公司 | Big data-based communication information automatic analysis system and equipment |
-
2021
- 2021-04-23 CN CN202110442038.4A patent/CN113051482A/en not_active Withdrawn
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114861019A (en) * | 2022-05-03 | 2022-08-05 | 北京博智瑞成科技有限公司 | Big data-based communication information automatic analysis system and equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108959270B (en) | Entity linking method based on deep learning | |
CN101320375B (en) | Digital book search method based on user click action | |
CN103914478B (en) | Webpage training method and system, webpage Forecasting Methodology and system | |
CN107862022B (en) | Culture resource recommendation system | |
CN109271477A (en) | A kind of method and system by internet building taxonomy library | |
CN100545847C (en) | A kind of method and system that blog articles is sorted | |
US20080294628A1 (en) | Ontology-content-based filtering method for personalized newspapers | |
CN108647322B (en) | Method for identifying similarity of mass Web text information based on word network | |
KR20150036117A (en) | Query expansion | |
CN108280114A (en) | A kind of user's literature reading interest analysis method based on deep learning | |
CN110222260A (en) | A kind of searching method, device and storage medium | |
CN107895303B (en) | Personalized recommendation method based on OCEAN model | |
CN112559684A (en) | Keyword extraction and information retrieval method | |
CN113282834A (en) | Web search intelligent ordering method, system and computer storage medium based on mobile internet data deep mining | |
CN111651584A (en) | Insurance article recommendation method based on user behavior characteristics and article attributes | |
CN102637179A (en) | Method and device for determining lexical item weighting functions and searching based on functions | |
JP2006318398A (en) | Vector generation method and device, information classifying method and device, and program, and computer readable storage medium with program stored therein | |
Reichling et al. | Matching human actors based on their texts: design and evaluation of an instance of the ExpertFinding framework | |
CN112669113A (en) | Product recommendation method and device, storage medium and electronic device | |
KR100973969B1 (en) | System for providing News Service that reduce Media-Bias Effect and Method therefor | |
US20090292691A1 (en) | System and Method for Building Multi-Concept Network Based on User's Web Usage Data | |
CN111696656A (en) | Doctor evaluation method and device of Internet medical platform | |
CN116610853A (en) | Search recommendation method, search recommendation system, computer device, and storage medium | |
CN113051482A (en) | Web page search intelligent matching recommendation method based on user feature recognition and behavior analysis | |
CN104615685B (en) | A kind of temperature evaluation method of network-oriented topic |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210629 |
|
WW01 | Invention patent application withdrawn after publication |