CN111400602A - Online learning system and method based on personalized recommendation - Google Patents

Online learning system and method based on personalized recommendation Download PDF

Info

Publication number
CN111400602A
CN111400602A CN202010199530.9A CN202010199530A CN111400602A CN 111400602 A CN111400602 A CN 111400602A CN 202010199530 A CN202010199530 A CN 202010199530A CN 111400602 A CN111400602 A CN 111400602A
Authority
CN
China
Prior art keywords
user
recommendation
course
learning
personalized recommendation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010199530.9A
Other languages
Chinese (zh)
Inventor
陈家峰
李书兵
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.)
Zhuhai Dulang Online Education Co ltd
Original Assignee
Zhuhai Dulang Online Education Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhuhai Dulang Online Education Co ltd filed Critical Zhuhai Dulang Online Education Co ltd
Priority to CN202010199530.9A priority Critical patent/CN111400602A/en
Publication of CN111400602A publication Critical patent/CN111400602A/en
Pending legal-status Critical Current

Links

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
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/3332—Query translation
    • G06F16/3334—Selection or weighting of terms from queries, including natural language queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06F16/335—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
    • G09B5/00—Electrically-operated educational appliances
    • G09B5/02—Electrically-operated educational appliances with visual presentation of the material to be studied, e.g. using film strip

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Business, Economics & Management (AREA)
  • Educational Administration (AREA)
  • Educational Technology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses an online learning system and method based on personalized recommendation, which comprises the following steps: the method aims at the problems that the increase of learning resources in an online learning system, the overload of the learning resources and the mass of the learning resources cause great troubles to users, and the users often need to spend a great deal of time and energy to search the needed learning resources. Designing and developing an online learning system based on personalized recommendation, and performing linear weighted fusion on a probability matrix decomposition algorithm based on fusion deep learning and a collaborative filtering recommendation algorithm based on a user through a coarse-grained weight calculation method to obtain a final personalized recommendation candidate set. The system can not only provide a platform for the user to learn anytime and anywhere, but also recommend learning resources interested by the user according to the interest characteristics of the user, effectively solve the problem of overload of the learning resources in the online learning system, and improve the experience effect and learning efficiency of the user.

Description

Online learning system and method based on personalized recommendation
Technical Field
The invention belongs to the technical field of online learning, and particularly relates to an online learning system and method based on personalized recommendation.
Background
With the popularization of internet life style and the innovation of learning style, online learning of the concept is getting hot and fast. The online learning is not limited by factors such as regions and time, all learning resources are shared, and the condition that the learning resources in different regions are unbalanced is effectively alleviated. The online learning system provides rich learning resources for users and brings trouble to the users. With the increase of learning resources in the system, the problem of learning resource overload gradually appears. The massive learning resources cause great troubles to users, and the users often need to spend a great deal of time and energy to search the needed learning resources. Therefore, how to enable the user to quickly find the needed resources in the massive learning resources becomes a problem to be solved urgently by the current online learning system. Therefore, the online learning system based on personalized recommendation is designed and developed, a platform for learning anytime and anywhere can be provided for the user, interested learning resources can be recommended according to the interest characteristics of the user, the problem of overload of the learning resources in the online learning system is effectively solved, and the experience effect and the learning efficiency of the user are improved.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an online learning system and method based on personalized recommendation.
The invention provides an online learning system based on personalized recommendation, which is characterized by comprising a view module, a controller module and a service module. The view module is a computer and a learning panel, is responsible for processing input information of a user, outputting and displaying the input information to the user, and is connected with the controller module through a network. The controller module is used for combining the data of the view module and the data of the service module and processing the information. The service module is responsible for operations such as service processing, data modification, storage and query. The data access function can convert the data transmitted by the controller into an expression mode which can be identified by the data, and access of the database is realized.
The online learning method based on the personalized recommendation mainly comprises the following steps:
step 1, a user performs online learning through a view module, a controller module decomposes a recommendation model according to historical data of the user in a service module and a probability matrix of deep learning, calculates the user's preference for courses, and acquires a recommendation candidate set;
step 2, the controller module calculates the course likeness of the user according to a collaborative filtering recommendation algorithm based on the user, and acquires a recommendation candidate set;
step 3, the controller module adopts a coarse-grained weight calculation method to obtain linear weighting weights corresponding to the two methods obtained in the steps 1 and 2;
and 4, the controller module performs linear weighted fusion on the recommendation candidate sets obtained by the two methods according to the corresponding weights to obtain a final personalized recommendation candidate set, stores the final personalized recommendation candidate set in the service module, and displays the final personalized recommendation candidate set to the user through a computer and a learning panel of the view module.
Preferably, the method for acquiring the recommended candidate set in step 1 is as follows:
and taking the user characteristic vector and the course characteristic vector as input, calculating the course preference of the user, and then selecting the course with the highest user preference. According to the selected courses, obtaining implicit characteristic vectors of all the courses in the category, and then calculating the user's preference of each course in the category according to the formula:
Figure BDA0002418874360000021
wherein u is a user feature vector, v is a course feature vector, unIs a feature vector of user n, vnIs the feature vector for course n.
And sorting the calculated likelihoods in a descending order, and selecting the courses with the top 30 likelihoods as a recommended candidate set for each user.
Preferably, the method for obtaining the recommended candidate set in step 2 is as follows:
taking the user feature vectors as input, and then calculating the cosine similarity between the user feature vectors according to the formula:
Figure BDA0002418874360000031
wherein u isiIs the feature vector of user i, ujIs the feature vector of user j.
And performing descending sorting on the cosine similarity obtained by calculation, and selecting the user with the top 30 cosine similarities as a similar user group for each user. Acquiring historical grading courses of similar user groups, and calculating the user preference of the similar user historical grading courses according to the formula:
Figure BDA0002418874360000032
where K (i) represents the set of users who scored course i, S (u, N) represents the N users most similar to user u, where N equals 30,
Figure BDA0002418874360000033
average score for all scored courses for user u, rviScoring lesson i for user v.
And then sorting in descending order according to the preference.
Preferably, the linear weighting method in step 3 is as follows:
and respectively calculating the prediction preference of each user to the self scored courses according to a probability matrix decomposition recommendation model integrating deep learning and a collaborative filtering recommendation algorithm based on the users. And constructing an objective function according to the predicted preference degree, and solving weight values corresponding to different methods by minimizing the objective function. The meaning is to minimize the sum of the squares of the actual likeness and predicted likeness errors, with an objective function of:
Figure BDA0002418874360000034
α therein1+α2＝1，0≤α1≤1，0≤α2≤1，α1Corresponding to probability matrix decomposition recommendation model representing fusion deep learningWeight value, α2Representing the weight values corresponding to the user-based collaborative filtering recommendation algorithm,
Figure BDA0002418874360000041
and expressing the predicted preference of the user i to the self scored course j, which is obtained by calculation according to the probability matrix decomposition recommendation model integrating deep learning.
Figure BDA0002418874360000042
Representing the predicted love degree of the user i to the self-scored course j calculated according to the collaborative filtering recommendation algorithm based on the user, RijRepresenting the user i's preference for his own scored course j.
Preferably, the linear weighted fusion method in step 4 is as follows:
and fusing the two recommendation candidate sets according to the formula to obtain a fused recommendation candidate set:
Figure BDA0002418874360000043
wherein IijShowing the preference of the user i to the course j after the fusion.
And sorting the fused recommendation candidate sets in a descending order according to the likelihoods, selecting courses with the top 10 likelihoods for each user as a final personalized recommendation candidate set, and storing the personalized recommendation candidate sets into a personalized recommendation database.
Compared with the prior art, the invention has the beneficial technical effects that: adding personalized recommendation into a traditional online learning system can enable a user to have greater pertinence and activity in the learning process, designing and realizing the online learning system based on a recommendation model, a course recommendation engine and specific service requirements, and performing linear weighted fusion on a probability matrix decomposition algorithm based on fusion deep learning and a collaborative filtering recommendation algorithm based on the user through a coarse-grained weight calculation method to obtain a final personalized recommendation candidate set; the system can construct a user model according to the user behavior, so that learning resources which the user may like are recommended to the user in a targeted manner, learning enthusiasm of the user is stimulated, and learning enthusiasm of the user is improved.
The attached drawings show
FIG. 1 is a flow chart of the steps of the present invention.
Fig. 2 is a structural frame diagram of the present invention.
Detailed Description
The online learning system based on personalized recommendation is characterized by comprising a view module, a controller module and a business module. The view module is a computer and a learning panel, is responsible for processing input information of a user, outputting and displaying the input information to the user, and is connected with the controller module through a network. The controller module is used for combining the data of the view module and the data of the service module and processing the information. The service module is responsible for operations such as service processing, data modification, storage and query. The data access function can convert the data transmitted by the controller into an expression mode which can be identified by the data, and access of the database is realized.
The online learning method based on the personalized recommendation mainly comprises the following steps:
step 1, a user performs online learning through a view module, a controller module decomposes a recommendation model according to historical data of the user in a service module and a probability matrix of deep learning, calculates the user's preference for courses, and acquires a recommendation candidate set;
specifically, the user characteristic vector and the course characteristic vector are used as input, the course preference of the user is calculated, and then the course with the highest user preference is selected. According to the selected courses, obtaining implicit characteristic vectors of all the courses in the category, and then calculating the user's preference of each course in the category according to the formula:
Figure BDA0002418874360000051
wherein u is a user feature vector, v is a course feature vector, unIs a feature vector of user n, vnIs the feature vector for course n.
And sorting the calculated likelihoods in a descending order, and selecting the courses with the top 30 likelihoods as a recommended candidate set for each user.
Step 2, the controller module calculates the course likeness of the user according to a collaborative filtering recommendation algorithm based on the user, and acquires a recommendation candidate set;
specifically, the user feature vectors are used as input, and then the cosine similarity between the user feature vectors is calculated according to the formula:
Figure BDA0002418874360000061
wherein u isiIs the feature vector of user i, ujIs the feature vector of user j.
And performing descending sorting on the cosine similarity obtained by calculation, and selecting the user with the top 30 cosine similarities as a similar user group for each user. Acquiring historical grading courses of similar user groups, and calculating the user preference of the similar user historical grading courses according to the formula:
Figure BDA0002418874360000062
where K (i) represents the set of users who scored course i, S (u, N) represents the N users most similar to user u, where N equals 30,
Figure BDA0002418874360000063
average score for all scored courses for user u, rviScoring lesson i for user v.
And then sorting in descending order according to the preference.
Step 3, the controller module adopts a coarse-grained weight calculation method to obtain linear weighting weights corresponding to the two methods obtained in the steps 1 and 2;
specifically, according to a probability matrix decomposition recommendation model integrating deep learning and a collaborative filtering recommendation algorithm based on users, the prediction preference of each user for the self-scored courses is calculated respectively. And constructing an objective function according to the predicted preference degree, and solving weight values corresponding to different methods by minimizing the objective function. The meaning is to minimize the sum of the squares of the actual likeness and predicted likeness errors, with an objective function of:
Figure BDA0002418874360000064
α therein1+α2＝1，0≤α1≤1，0≤α2≤1，α1Representing weight values corresponding to the probability matrix decomposition recommendation model for fusion deep learning, α2Representing the weight values corresponding to the user-based collaborative filtering recommendation algorithm,
Figure BDA0002418874360000071
and expressing the predicted preference of the user i to the self scored course j, which is obtained by calculation according to the probability matrix decomposition recommendation model integrating deep learning.
Figure BDA0002418874360000072
Representing the predicted love degree of the user i to the self-scored course j calculated according to the collaborative filtering recommendation algorithm based on the user, RijRepresenting the user i's preference for the self-scored course j.
And 4, the controller module performs linear weighted fusion on the recommendation candidate sets obtained by the two methods according to the corresponding weights to obtain a final personalized recommendation candidate set, stores the final personalized recommendation candidate set in the service module, and displays the final personalized recommendation candidate set to the user through a computer and a learning panel of the view module.
Specifically, the two recommendation candidate sets are fused according to the formula to obtain a fused recommendation candidate set:
Figure BDA0002418874360000073
wherein IijShowing the preference of the user i to the course j after the fusion.
And sorting the fused recommendation candidate sets in a descending order according to the likelihoods, selecting courses with the top 10 likelihoods for each user as a final personalized recommendation candidate set, and storing the personalized recommendation candidate sets into a personalized recommendation database.
The present invention has been described in detail, and the principle and embodiments of the present invention are explained by using specific examples, which are only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present disclosure should not be construed as a limitation to the present invention.

Claims (6)

1. An online learning system based on personalized recommendation is characterized by comprising a view module, a controller module and a service module; the view module is a computer and a learning panel, is responsible for processing input information of a user, outputting and displaying the input information to the user, and is connected with the controller module through a network; the controller module is used for combining the data of the view module and the data of the service module and processing the information, and the service module is responsible for the operations of service processing, data modification, storage, query and the like; the data access function can convert the data transmitted by the controller into an expression mode which can be identified by the data, and access of the database is realized.
2. The online learning method based on the personalized recommendation mainly comprises the following steps:
step 1, a user performs online learning through a view module, a controller module decomposes a recommendation model according to historical data of the user in a service module and a probability matrix of deep learning, calculates the user's preference for courses, and acquires a recommendation candidate set;
step 2, the controller module calculates the course likeness of the user according to a collaborative filtering recommendation algorithm based on the user, and acquires a recommendation candidate set;
step 3, the controller module adopts a coarse-grained weight calculation method to obtain linear weighting weights corresponding to the two methods obtained in the steps 1 and 2;
and 4, the controller module performs linear weighted fusion on the recommendation candidate sets obtained by the two methods according to the corresponding weights to obtain a final personalized recommendation candidate set, stores the final personalized recommendation candidate set in the service module, and displays the final personalized recommendation candidate set to the user through a computer and a learning panel of the view module.
3. The online learning method based on personalized recommendation of claim 2,
the method for acquiring the recommended candidate set in the step 1 comprises the following steps: the method comprises the following steps of taking a user characteristic vector and a course characteristic vector as input, calculating the degree of love of a user to a course, then selecting the course with the highest degree of love of the user, acquiring implicit characteristic vectors of all courses under the category according to the selected course, and then calculating the degree of love of the user to each course under the category according to a formula:
Figure FDA0002418874350000011
wherein u is a user feature vector, v is a course feature vector, unIs a feature vector of user n, vnAnd sorting the calculated likelihoods in a descending order for the feature vectors of the courses n, and selecting the courses with the top 30 likelihoods as a recommended candidate set for each user.
4. The online learning method based on personalized recommendation of claim 2, wherein:
the method for acquiring the recommended candidate set in the step 2 is as follows: taking the user feature vectors as input, and then calculating the cosine similarity between the user feature vectors according to the formula:
Figure FDA0002418874350000021
wherein u isiIs the feature vector of user i, ujA feature vector for user j; sorting the cosine similarity obtained by calculation in a descending order, and selecting the cosine similarity of the top 30 for each userUsers are taken as similar user groups; acquiring historical grading courses of similar user groups, and calculating the user preference of the similar user historical grading courses according to the formula:
Figure FDA0002418874350000022
where K (i) represents the set of users who scored course i, S (u, N) represents the N users most similar to user u, where N equals 30,
Figure FDA0002418874350000023
average score for all scored courses for user u, rviScoring course i for user v; and then sorting in descending order according to the preference.
5. The online learning method based on personalized recommendation of claim 2, wherein:
the linear weighting method in step 3 is: according to a probability matrix decomposition recommendation model integrating deep learning and a collaborative filtering recommendation algorithm based on users, the prediction preference of each user for the self scored courses is respectively calculated, an objective function is constructed according to the prediction preference, and the weight values corresponding to different methods are solved by minimizing the objective function; the meaning is to minimize the sum of the squares of the actual likeness and predicted likeness errors, with an objective function of:
Figure FDA0002418874350000031
α therein1+α2＝1，0≤α1≤1，0≤α2≤1，α1Representing weight values corresponding to the probability matrix decomposition recommendation model for fusion deep learning, α2Representing the weight values corresponding to the user-based collaborative filtering recommendation algorithm,
Figure FDA0002418874350000032
representing probability matrices learned from fusion depthThe user i obtains the predicted love degree of the graded course j by the decomposition recommendation model,
Figure FDA0002418874350000033
representing the predicted love degree of the user i to the self-scored course j calculated according to the collaborative filtering recommendation algorithm based on the user, RijRepresenting the user i's preference for his own scored course j.
6. The online learning method based on personalized recommendation of claim 2, wherein:
the linear weighting fusion mode in the step 4 is as follows: and fusing the two recommendation candidate sets according to the formula to obtain a fused recommendation candidate set:
Figure FDA0002418874350000034
wherein IijShowing the preference degree of the user i to the course j after the fusion; and sorting the fused recommendation candidate sets in a descending order according to the likelihoods, selecting courses with the top 10 likelihoods for each user as a final personalized recommendation candidate set, and storing the personalized recommendation candidate sets into a personalized recommendation database.
CN202010199530.9A 2020-03-20 2020-03-20 Online learning system and method based on personalized recommendation Pending CN111400602A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010199530.9A CN111400602A (en) 2020-03-20 2020-03-20 Online learning system and method based on personalized recommendation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010199530.9A CN111400602A (en) 2020-03-20 2020-03-20 Online learning system and method based on personalized recommendation

Publications (1)

Publication Number Publication Date
CN111400602A true CN111400602A (en) 2020-07-10

Family

ID=71428976

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010199530.9A Pending CN111400602A (en) 2020-03-20 2020-03-20 Online learning system and method based on personalized recommendation

Country Status (1)

Country Link
CN (1) CN111400602A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112733036A (en) * 2021-01-21 2021-04-30 唐亮 Knowledge point recommendation method and device, storage medium and electronic device

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110178964A1 (en) * 2010-01-21 2011-07-21 National Cheng Kung University Recommendation System Using Rough-Set and Multiple Features Mining Integrally and Method Thereof
CN106600302A (en) * 2015-10-19 2017-04-26 玺阅信息科技（上海）有限公司 Hadoop-based commodity recommendation system
CN107436950A (en) * 2017-08-07 2017-12-05 苏州大学 A kind of itinerary recommends method and system
CN108492230A (en) * 2018-04-03 2018-09-04 四川长虹电器股份有限公司 The system and method for internet service resource popularization based on distributed structure/architecture
CN109241405A (en) * 2018-08-13 2019-01-18 华中师范大学 A kind of associated education resource collaborative filtering recommending method of knowledge based and system
CN109684538A (en) * 2018-12-03 2019-04-26 重庆邮电大学 A kind of recommended method and recommender system based on individual subscriber feature
CN109919810A (en) * 2019-01-22 2019-06-21 山东科技大学 Student's modeling and personalized course recommended method in on-line study system
CN110399549A (en) * 2018-04-24 2019-11-01 河南理工大学 A kind of collaborative filtering method of user oriented interest reconciliation similarity
CN110807154A (en) * 2019-11-08 2020-02-18 内蒙古工业大学 Recommendation method and system based on hybrid deep learning model

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110178964A1 (en) * 2010-01-21 2011-07-21 National Cheng Kung University Recommendation System Using Rough-Set and Multiple Features Mining Integrally and Method Thereof
CN106600302A (en) * 2015-10-19 2017-04-26 玺阅信息科技（上海）有限公司 Hadoop-based commodity recommendation system
CN107436950A (en) * 2017-08-07 2017-12-05 苏州大学 A kind of itinerary recommends method and system
CN108492230A (en) * 2018-04-03 2018-09-04 四川长虹电器股份有限公司 The system and method for internet service resource popularization based on distributed structure/architecture
CN110399549A (en) * 2018-04-24 2019-11-01 河南理工大学 A kind of collaborative filtering method of user oriented interest reconciliation similarity
CN109241405A (en) * 2018-08-13 2019-01-18 华中师范大学 A kind of associated education resource collaborative filtering recommending method of knowledge based and system
CN109684538A (en) * 2018-12-03 2019-04-26 重庆邮电大学 A kind of recommended method and recommender system based on individual subscriber feature
CN109919810A (en) * 2019-01-22 2019-06-21 山东科技大学 Student's modeling and personalized course recommended method in on-line study system
CN110807154A (en) * 2019-11-08 2020-02-18 内蒙古工业大学 Recommendation method and system based on hybrid deep learning model

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
巩晓悦: "基于个性化推荐的在线学习系统研究与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112733036A (en) * 2021-01-21 2021-04-30 唐亮 Knowledge point recommendation method and device, storage medium and electronic device

Similar Documents

Publication Publication Date Title
CN107273490B (en) Combined wrong question recommendation method based on knowledge graph
CN111753060B (en) Information retrieval method, apparatus, device and computer readable storage medium
CN110555112B (en) Interest point recommendation method based on user positive and negative preference learning
CN105808590B (en) Search engine implementation method, searching method and device
CN112434517B (en) Community question-answering website answer ordering method and system combined with active learning
CN107122455A (en) A kind of network user's enhancing method for expressing based on microblogging
CN110795571A (en) Cultural tourism resource recommendation method based on deep learning and knowledge graph
CN113744032B (en) Book recommendation method, related device, equipment and storage medium
CN110110225A (en) Online education recommended models and construction method based on user behavior data analysis
CN106610970A (en) Collaborative filtering-based content recommendation system and method
CN112527993B (en) Cross-media hierarchical deep video question-answer reasoning framework
CN112132536A (en) Post recommendation method, system, computer equipment and storage medium
CN115964560B (en) Information recommendation method and equipment based on multi-mode pre-training model
Zhong et al. Design of a personalized recommendation system for learning resources based on collaborative filtering
CN113806630A (en) Attention-based multi-view feature fusion cross-domain recommendation method and device
CN113011172A (en) Text processing method and device, computer equipment and storage medium
CN107016566A (en) User model construction method based on body
CN113239209A (en) Knowledge graph personalized learning path recommendation method based on RankNet-transformer
CN109492085A (en) Method, apparatus, terminal and storage medium are determined based on the answer of data processing
CN114781503A (en) Click rate estimation method based on depth feature fusion
CN111400602A (en) Online learning system and method based on personalized recommendation
CN111949894B (en) Collaborative filtering personalized recommendation method based on multi-space interaction
CN113449182A (en) Knowledge information personalized recommendation method and system
CN116957128A (en) Service index prediction method, device, equipment and storage medium
CN110795640A (en) Adaptive group recommendation method for compensating group member difference

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20200710

RJ01 Rejection of invention patent application after publication