CN113139135A - Improved collaborative filtering network course recommendation algorithm - Google Patents
Improved collaborative filtering network course recommendation algorithm Download PDFInfo
- Publication number
- CN113139135A CN113139135A CN202110522520.9A CN202110522520A CN113139135A CN 113139135 A CN113139135 A CN 113139135A CN 202110522520 A CN202110522520 A CN 202110522520A CN 113139135 A CN113139135 A CN 113139135A
- Authority
- CN
- China
- Prior art keywords
- course
- user
- courses
- degree
- network
- 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.)
- Granted
Links
- 238000001914 filtration Methods 0.000 title claims abstract description 15
- 238000012163 sequencing technique Methods 0.000 claims abstract description 4
- 230000006399 behavior Effects 0.000 claims description 12
- 238000000034 method Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000011161 development Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010348 incorporation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 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
- G06F16/9536—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
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
- G06Q50/2053—Education institution selection, admissions, or financial aid
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Databases & Information Systems (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an improved collaborative filtering network course recommendation algorithm, which is characterized in that according to the completion degree, collection behavior and user evaluation feedback data of a user u on a known course i, the interest degree of the user u on the known course i is calculated, all network courses and the subject labels of each network course are obtained, and the network courses are divided into different sub-course data sets according to the subject labels of the network courses; setting an interest degree threshold, and when the interest degree of the user u for a certain course exceeds the threshold, determining that the user u has a tendency to the course; calculating the matching degree w between the known course i and other unknown courses j in the sub-course data setijTo obtain the interest r of the user u in the unknown course juj(ii) a Sequencing all unknown courses j in the sub-course data set according to the interestingness of the user u, and taking the top N courses as recommended courses; the network course recommendation algorithm can utilize the advantages of the mobile terminalAnd assisting in off-line classroom teaching.
Description
Technical Field
The invention belongs to the technical field of network course recommendation, and particularly relates to an improved collaborative filtering network course recommendation algorithm.
Background
With the development of networks and mobile terminals, mobile learning is becoming popular, and network learning inevitably causes some problems in a new era:
1) the campus classroom mode and the mobile network learning mode lack effective fusion.
The evolution of web3.0 has changed the traditional web learning environment. With the rapid development of mobile internet technology, online lectures are pursued by many people. In 2012, after the mocc (Massive Open Online Course) is started for a year, the mocc rapidly increases in temperature globally. However, up to now, the network classroom and the traditional classroom lack effective fusion and even contradict each other. How to complement the advantages of mobile learning and classroom learning so as to improve the efficiency of learners is a problem which needs to be solved urgently.
2) The environment of the network learning resource flooding presents a serious challenge to the network learning resource knowledge discovery.
Compared with the traditional teaching resources with relatively rare classroom and high quality, the online learning resources have wide and diversified sources, huge quantity and remarkable acceleration. The massive online learning resources provide massive information for learners and bring certain obstruction to resource retrieval of learners. If the network learning resources lack effective management, the utilization efficiency is very low, and the learners are submerged in the information sea.
3) A small range of collaborative learning lacks perfect platform support.
The current society is a society of 'human-human interconnection and resource sharing', and sharing and cooperation are inevitable development trends of learning. In classroom learning, it is often necessary to discuss problems and complete tasks in a group cooperation manner. Random distribution cannot make good use of the advantages and avoid the disadvantages, and the greatest advantages of each person are played. Free teams often lack mobility of personnel, creating more problems. In popular web learning, users mostly watch videos alone. Lack a mobile learning platform, gather the classmates of the same hobbies, produce the cooperative group, finish the common study task.
Disclosure of Invention
In order to overcome the defects in the prior art, the application provides an improved collaborative filtering network course recommendation algorithm, and the advantages of a mobile terminal are utilized to assist in off-line classroom teaching. Meanwhile, a collaborative learning group is created according to the learning subjects and interests of different classmates, learning resources are intelligently aggregated to members in the group, and the members in the group can conveniently study in a research and cooperative manner.
The technical scheme adopted by the invention is as follows:
an improved collaborative filtering network course recommendation algorithm comprises the following steps:
step 1: calculating the interest degree of the user u in the known course i according to the completion degree, the collection behavior and the feedback data of the user u on the known course i,
step 2: acquiring all network courses and subject labels of each network course, and dividing the network courses into different sub-course data sets according to the subject labels of the network courses;
and step 3: setting an interest degree threshold, and when the interest degree of the user u for a certain course exceeds the threshold, determining that the user u has a tendency to the course; calculating the matching degree w between the known course i and other unknown courses j in the sub-course data setijTo obtain the interest r of the user u in the unknown course juj；
And 4, step 4: and sequencing the interestingness of all unknown courses j in the sub-course data set according to the user u, wherein the top N courses are taken as recommended courses.
Further, the method for calculating the interest degree of the user u in the known course i in the step 1 comprises the following steps:
rui＝Tui(0.08Pui+0.6)+0.6Lui
wherein r isuiFor user u interest level, r, in known course iuiHigher values represent higher interest; puiScoring course i for user u; t isuiThe current completion degree of the course i for the user u; l isuiAnd judging whether the user u has collection behaviors for the course i, if so, the collection behaviors are 1, otherwise, the collection behaviors are 0.
Further, the interestingness threshold is taken to be 0.76.
Further, the method for predicting the unknown course j based on the interest degree of the known course i in the step 3 comprises the following steps:
step 3.1: firstly, calculating the matching degree w between the course i and the unknown course jij；
Step 3.2: obtaining the interest degree r of the user u to the unknown course j based on the corrected matching degreeuj。
Further, the degree of incorporation wijThe calculation method comprises the following steps:
wherein, wijRepresenting the matching degree between the i course and the j course, N (i) is the number of users having tendency to the i course, N (j) is the number of users having tendency to the j course, and n is the intersection;
further, in order to increase the recommended coverage rate, the weight of the hot course is punished, and the corrected fitness calculation formula is as follows:
wherein, n (u) is the number of courses that user u has a tendency. N (i) and n (j) are the number of users having a tendency to the i course and the i course, respectively.
Calculating interest degree P of user u in unknown course juj：
Wherein N (u) is a course set with user tendency, S (i, K) is a set of K items with highest matching degree with the course i, and wij' is the degree of match of courses i and j, ruiIs the interest level of user u in lesson i.
The invention has the beneficial effects that:
the application provides an improved collaborative filtering network course recommendation algorithm, which is used for assisting teaching in offline classes by utilizing the advantages of a mobile terminal. Meanwhile, a collaborative learning group is created according to the learning subjects and interests of different classmates, learning resources are intelligently aggregated to members in the group, and the members in the group can conveniently study in a research and cooperative manner.
Drawings
FIG. 1 is a flow chart of the network course recommendation algorithm of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
An improved collaborative filtering network course recommendation algorithm as shown in fig. 1 comprises the following steps:
step 1: calculating the interest degree of the user u in the known course i according to the completion degree, the collection behavior and the feedback data such as the user score of the user u on the known course i, wherein the calculation formula is as follows:
rui＝Tui(0.08Pui+0.6)+0.6Lui
wherein r isuiFor user u interest level, r, in known course iuiValue range [0,2 ]]The accuracy is to ten-thousandth, and the higher the numerical value is, the higher the interest degree is; puiThe score of the user u for the course i is given as [1,10 ]]By default, 7; t isuiThe current completion degree of the user u to the course i is taken as the value range [0,1 ]]To a percentile, LuiAnd judging whether the user u has collection behaviors for the course i, if so, the collection behaviors are 1, otherwise, the collection behaviors are 0.
Step 2: all network courses and the subject labels of each network course are obtained, and the network courses are divided into different sub-course data sets according to the subject labels of the network courses. For example, java programming language course, programming mode, C language programming and the like belong to computer technology labels and are divided into computer technology sub-course data sets; the labels are edited according to disciplinary classification.
And step 3: defining the interest degree r of the user u in the known course iui>0.76, the user u is considered to have a tendency to the course i.
Because the unknown course j and the course i belong to the same seed course data set; therefore, the prediction of the unknown course j in the same seed course data set is realized based on the interest degree of the known course i, and the prediction method comprises the following steps:
step 3.1: firstly, calculating the matching degree w between the course i and the unknown course jij：
Wherein, wijThe coordination degree between the i course and the j course is shown, N (i) is the number of users having tendency to the i course, N (j) is the number of users having tendency to the j course, and n is the intersection.
Step 3.2: to increase the coverage of recommendations, weights for hot courses are penalized. While penalizing the weight of an active User according to the iuf (inverse User frequency), a parameter that is the inverse of the log of User activity. The formula for calculating the corrected fitness is as follows:
wherein, n (u) is the number of courses that user u has a tendency. N (i) and n (j) are the number of users having a tendency to the i course and the i course, respectively.
Step 3.3: after the corrected matching degree is obtained, the interest degree P of the user u in the unknown course j is calculated through the following formulauj：
Wherein N (u) is a course set with user tendency, S (i, K) is a set of K items with highest matching degree with the course i, and wij' is lessonDegree of fit, r, of ranges i and juiIs the interest level of user u in lesson i. The implication of this formula is that lessons with a higher degree of engagement with lessons that the user has historically tended are more likely to get a higher ranking in the user's recommendation list.
And 4, step 4: sequencing all unknown courses j in the sub-course data set according to the interestingness of a user u, taking the top N courses as recommended courses, wherein the evaluation indexes such as coverage rate, accuracy and the like can be obviously influenced by the value of N, and finally determining a TopN list recommended to the user through a test experiment.
The method for finally determining the recommended course number N through the test experiment comprises the following steps: and selecting values for N from 1 to 28 at an interval of 3, wherein the obtained accuracy rate generally has a descending trend after ascending, when the value of N is 16, the accuracy rate is the highest, and the accuracy rate becomes a descending trend along with the increase of the value of N.
The above embodiments are only used for illustrating the design idea and features of the present invention, and the purpose of the present invention is to enable those skilled in the art to understand the content of the present invention and implement the present invention accordingly, and the protection scope of the present invention is not limited to the above embodiments. Therefore, all equivalent changes and modifications made in accordance with the principles and concepts disclosed herein are intended to be included within the scope of the present invention.
Claims (7)
1. An improved collaborative filtering network course recommendation algorithm is characterized by comprising the following steps:
step 1: calculating the interest degree of the user u in the known course i according to the completion degree, the collection behavior and the feedback data of the user u on the known course i,
step 2: acquiring all network courses and subject labels of each network course, and dividing the network courses into different sub-course data sets according to the subject labels of the network courses;
and step 3: setting an interest degree threshold, and when the interest degree of the user u for a certain course exceeds the threshold, determining that the user u has a tendency to the course; calculating the matching degree w between the known course i and other unknown courses j in the sub-course data setijTo obtain the interest r of the user u in the unknown course juj；
And 4, step 4: and sequencing the interestingness of all unknown courses j in the sub-course data set according to the user u, wherein the top N courses are taken as recommended courses.
2. The improved collaborative filtering network course recommendation algorithm as claimed in claim 1, wherein the method for calculating the interest degree of the user u in the known course i in step 1 comprises:
rui＝Tui(0.08Pui+0.6)+0.6Lui
wherein r isuiFor user u interest level, r, in known course iuiHigher values represent higher interest; puiScoring course i for user u; t isuiThe current completion degree of the course i for the user u; l isuiAnd judging whether the user u has collection behaviors for the course i, if so, the collection behaviors are 1, otherwise, the collection behaviors are 0.
3. The improved collaborative filtering network course recommendation algorithm of claim 1, wherein the interestingness threshold is 0.76.
4. The improved collaborative filtering network course recommendation algorithm as claimed in claim 1, wherein the method for implementing prediction of unknown course j based on interest of known course i in step 3 is as follows:
step 3.1: firstly, calculating the matching degree w between the course i and the unknown course jij；
Step 3.2: obtaining the interest degree r of the user u to the unknown course j based on the corrected matching degreeuj。
5. The improved collaborative filtering web course recommendation algorithm according to claim 4, wherein the degree of compliance wijThe calculation method comprises the following steps:
wherein, wijThe coordination degree between the i course and the j course is shown, N (i) is the number of users having tendency to the i course, N (j) is the number of users having tendency to the j course, and n is the intersection.
6. The improved collaborative filtering network course recommendation algorithm according to claim 5, wherein to increase the coverage of recommendation, penalize the weight of hot courses, the modified fitness calculation formula is:
wherein, n (u) is the number of courses that user u has a tendency. N (i) and n (j) are the number of users having a tendency to the i course and the i course, respectively.
7. The improved collaborative filtering network course recommendation algorithm as claimed in claim 4, wherein the interest level P of the user u in the unknown course j is calculateduj：
Wherein N (u) is a course set with user tendency, S (i, K) is a set of K items with highest matching degree with the course i, and wij' is the degree of match of courses i and j, ruiIs the interest level of user u in lesson i.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110522520.9A CN113139135B (en) | 2021-05-13 | 2021-05-13 | Improved collaborative filtering network course recommendation algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110522520.9A CN113139135B (en) | 2021-05-13 | 2021-05-13 | Improved collaborative filtering network course recommendation algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113139135A true CN113139135A (en) | 2021-07-20 |
| CN113139135B CN113139135B (en) | 2023-09-19 |
Family
ID=76817396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110522520.9A Active CN113139135B (en) | 2021-05-13 | 2021-05-13 | Improved collaborative filtering network course recommendation algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113139135B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110208822A1 (en) * | 2010-02-22 | 2011-08-25 | Yogesh Chunilal Rathod | Method and system for customized, contextual, dynamic and unified communication, zero click advertisement and prospective customers search engine |
| CN103167023A (en) * | 2013-02-05 | 2013-06-19 | 广东工业大学 | Architecture professional study mobile platform based on cloud computing and location service |
| CN105786983A (en) * | 2016-02-15 | 2016-07-20 | 云南电网有限责任公司 | Employee individualized-learning recommendation method based on learning map and collaborative filtering |
| CN108510307A (en) * | 2018-02-25 | 2018-09-07 | 心触动（武汉）科技有限公司 | A kind of course recommendation method and system |
| 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 |
| CN109299372A (en) * | 2018-10-18 | 2019-02-01 | 浙江正元智慧科技股份有限公司 | A kind of wisdom curricula-variable recommended method based on collaborative filtering |
| CN109472286A (en) * | 2018-09-30 | 2019-03-15 | 浙江工业大学 | Books in University Library recommended method based on interest-degree model Yu the type factor |
| CN109801197A (en) * | 2019-01-09 | 2019-05-24 | 东华大学 | A kind of educational administration's assistant system based on Collaborative Filtering Recommendation Algorithm |
| CN110134871A (en) * | 2019-05-27 | 2019-08-16 | 湖南大学 | A kind of dynamic course recommended method based on course and learner's network structure |
| CN110659423A (en) * | 2019-09-19 | 2020-01-07 | 辽宁工程技术大学 | School side learning material recommendation method based on collaborative filtering |
| US20200042648A1 (en) * | 2018-08-02 | 2020-02-06 | Microsoft Technology Licensing, Llc | Recommending development tool extensions based on media type |
| CN111143750A (en) * | 2020-01-06 | 2020-05-12 | 成都信息工程大学 | University computer class course management platform based on fragmented learning |
| CN112597392A (en) * | 2020-12-25 | 2021-04-02 | 厦门大学 | Recommendation system based on dynamic attention and hierarchical reinforcement learning |
-
2021
- 2021-05-13 CN CN202110522520.9A patent/CN113139135B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110208822A1 (en) * | 2010-02-22 | 2011-08-25 | Yogesh Chunilal Rathod | Method and system for customized, contextual, dynamic and unified communication, zero click advertisement and prospective customers search engine |
| CN103167023A (en) * | 2013-02-05 | 2013-06-19 | 广东工业大学 | Architecture professional study mobile platform based on cloud computing and location service |
| CN105786983A (en) * | 2016-02-15 | 2016-07-20 | 云南电网有限责任公司 | Employee individualized-learning recommendation method based on learning map and collaborative filtering |
| CN108510307A (en) * | 2018-02-25 | 2018-09-07 | 心触动（武汉）科技有限公司 | A kind of course recommendation method and system |
| US20200042648A1 (en) * | 2018-08-02 | 2020-02-06 | Microsoft Technology Licensing, Llc | Recommending development tool extensions based on media type |
| 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 |
| CN109472286A (en) * | 2018-09-30 | 2019-03-15 | 浙江工业大学 | Books in University Library recommended method based on interest-degree model Yu the type factor |
| CN109299372A (en) * | 2018-10-18 | 2019-02-01 | 浙江正元智慧科技股份有限公司 | A kind of wisdom curricula-variable recommended method based on collaborative filtering |
| CN109801197A (en) * | 2019-01-09 | 2019-05-24 | 东华大学 | A kind of educational administration's assistant system based on Collaborative Filtering Recommendation Algorithm |
| CN110134871A (en) * | 2019-05-27 | 2019-08-16 | 湖南大学 | A kind of dynamic course recommended method based on course and learner's network structure |
| CN110659423A (en) * | 2019-09-19 | 2020-01-07 | 辽宁工程技术大学 | School side learning material recommendation method based on collaborative filtering |
| CN111143750A (en) * | 2020-01-06 | 2020-05-12 | 成都信息工程大学 | University computer class course management platform based on fragmented learning |
| CN112597392A (en) * | 2020-12-25 | 2021-04-02 | 厦门大学 | Recommendation system based on dynamic attention and hierarchical reinforcement learning |
Non-Patent Citations (2)
| Title |
|---|
| 尚立: "基于协同过滤的改进课程推荐算法", 《科技传播》, pages 132 - 134 * |
| 高立强等: "改进型协同过滤的网络课程推荐算法", 电脑与电信, pages 53 - 56 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113139135B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112149001B (en) | Learning companion recommendation system and method | |
| Murata et al. | Link prediction of social networks based on weighted proximity measures | |
| CN110442790A (en) | Recommend method, apparatus, server and the storage medium of multi-medium data | |
| CN104133817A (en) | Online community interaction method and device and online community platform | |
| CN111540244A (en) | Multi-terminal online teaching system based on big data analysis | |
| CN109492076B (en) | Community question-answer website answer credible evaluation method based on network | |
| Skopik et al. | Start trusting strangers? bootstrapping and prediction of trust | |
| CN111723973B (en) | Learning effect optimization method based on user behavior causal relationship in MOOC log data | |
| CN110134871B (en) | Dynamic course recommendation method based on course and learner network structure | |
| Li et al. | Multi-objective optimization-based recommendation for massive online learning resources | |
| CN110807509A (en) | Depth knowledge tracking method based on Bayesian neural network | |
| Su et al. | Next check-in location prediction via footprints and friendship on location-based social networks | |
| Wei et al. | Cooperation dynamics on collaborative social networks of heterogeneous population | |
| Tang et al. | A literature review of personalized learning algorithm | |
| CN106407379A (en) | Hadoop platform based movie recommendation method | |
| CN110765362A (en) | Collaborative filtering personalized learning recommendation method based on learning condition similarity | |
| Guo et al. | A large-scale group decision-making method fusing three-way clustering and regret theory under fuzzy preference relations | |
| Lin | Learning information recommendation based on text vector model and support vector machine | |
| CN112712449A (en) | Vocational education online teaching system based on artificial intelligence and big data | |
| CN113139135A (en) | Improved collaborative filtering network course recommendation algorithm | |
| Yao et al. | [Retracted] Prediction of College Students’ Employment Rate Based on Gray System | |
| CN111582595B (en) | Emergency decision method and device for natural disasters and electronic equipment | |
| Li et al. | A Network Public Opinion Trend Estimation Model Using a Scale‐Free Network Algorithm | |
| CN111079003A (en) | Technical scheme of potential preference correlation prediction model with social circle as key support | |
| Li et al. | Design of Intelligent Push System for Teaching Resources Based on User Profile |
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 |