CN106528656B - A kind of method and system for realizing that course is recommended based on student's history and real-time learning state parameter - Google Patents
A kind of method and system for realizing that course is recommended based on student's history and real-time learning state parameter Download PDFInfo
- Publication number
- CN106528656B CN106528656B CN201610911525.XA CN201610911525A CN106528656B CN 106528656 B CN106528656 B CN 106528656B CN 201610911525 A CN201610911525 A CN 201610911525A CN 106528656 B CN106528656 B CN 106528656B
- Authority
- CN
- China
- Prior art keywords
- course
- resources
- student
- standard
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000013439 planning Methods 0.000 claims description 49
- 230000000875 corresponding effect Effects 0.000 claims description 28
- 230000002596 correlated effect Effects 0.000 claims description 17
- 238000004458 analytical method Methods 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000007726 management method Methods 0.000 claims description 8
- 238000013139 quantization Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 239000000463 material Substances 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000003252 repetitive effect Effects 0.000 description 2
- 238000013517 stratification Methods 0.000 description 2
- 230000009885 systemic effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24578—Query processing with adaptation to user needs using ranking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- 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
- 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—Systems or methods specially adapted for 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
Abstract
The present invention provides the course resources recommended methods and system suitable for E-learning Platform of a kind of specialization, from on general or other fields information recommendation method and system it is different, the present invention forms tree-shaped data structure to the magnanimity course resources on E-learning Platform, and the school work feature for comprehensively considering professional knowledge system and magnanimity student forms user's portrait model, realize that classification mode identifies using the study history and learning state of individual student on this basis, and then the degree of association of the classification according to belonging to student and course, personalized course is generated in big data course resources library to recommend.
Description
Technical field
The present invention relates to field of computer technology, more particularly to one kind are real based on student's history and real-time learning state parameter
The method and system that existing course is recommended.
Background technique
Web education is of great significance to construction learning society, promotion greening to all citizens, especially leads towards profession
The Web education in domain is conducive to the wide-scale distribution of high-grade, precision and advanced knowledge, and the personnel training for promoting every profession and trade is horizontal, reduces society and enterprise
Industry training cost alleviates the imbalance of region educational development.Thus, the exploitation and application of Network Education System are computer technologies
The important directions of development.
Through retrieving, the technological improvement of Network Education System is mainly reflected in the prior art: establishing education dependency number
According to storage, transmission and acquisition basic network platform；Realize the interaction in learning process；And to study permission and progress
Monitoring, record, feedback and management etc..
For example, application No. is 201110263215.9 Chinese invention patent application " network curriculum learning platform and communications
In method ", which includes WEB server, logical server, database server, content integrated service
Device, content distribution server, material server, streaming media server and network courses packet server, network courses packet server
It is stored with e-learning content packet, e-learning content packet and WEB server interact communication, net by application programming interfaces
Network course learning platform reads user basic information by the application programming interfaces in WEB server and gives e-learning content packet,
And receive user's dynamic learning information of e-learning content packet transmission.The learning functionality that the platform is realized includes setting study
Permission, the publication of the learning Content page, video online request, learning information inquiry and learning information feedback.Student is terminating to learn
When, Network Learning Platform records this study situation of student, including which resource, study to which chapters and sections, chapters and sections browsed
Test is by state etc., and the platform provides corresponding prompt and navigation information on the learning Content page when student logs in next time.
Application No. is 201610374875.7 Chinese invention patent application " a kind of network-based interactive education system and
In application method ", the application terminals such as tablet computer, smart phone, PC are divided into teacher's terminal, student terminal and parent's terminal, with
Upper terminal connects mutual education Cloud Server by internet respectively, and mutual education Cloud Server stores teaching resource exam pool, can
It is transferred for teaching, carries out information push using intelligent terminal tool, and can realize interactive answer, and carry out the real-time of data
Statistics.
It includes video, audio, written historical materials, multimedia courseware, study that Network Education System, which can be collected, integrate, issuing,
The course resources of the diversified forms such as software.The data of magnanimity are formd with the extreme enrichment of course resources in big data era
On the one hand resources bank establishes good basis with shared for the abundant propagation of knowledge information, on the other hand also fits to student
When course planning with selection bring difficulty.
In the massive information epoch, the information recommendation based on content is to aid in user and resource is overcome to spread unchecked bring selection barrier
Hinder, a kind of effective data that quickly positioning meets the resource of self-demand are analyzed and technology of sharing.For example, application number
In 201010273633.1 Chinese invention patent application " a kind of recommended method and system of product information ", user is predefined
Recommended products collection and/or product recommended products collection；The network operation for obtaining the first user is grasped according to the network of the first user
Make to determine Products Show type；According to determining Products Show type, from the recommended products collection and/or the network of the first user
The recommended products for operating associated first product, which is concentrated, determines that required under corresponding Products Show type is that the first user recommends
Product information.This method and system can be more accurate the product information that may need of determination user.
Application No. is the Chinese invention patent application of 201210012345.X " information recommendation methods based on potential community "
It is related to a kind of information recommendation method based on potential community.User is excavated from user's history record by feature discovery technique
Interest model, and potential community discovery is carried out according to the model and excavates the hierarchical relationship between community；Then, it is mentioned by feature
The characteristics of taking and extract interest set that may be present from object to be recommended, and combining community hierarchical structure, quickly navigates to
Community to be recommended；It is pushed away finally, calculating the similarity between community to be recommended and object interest set to be recommended to determine whether meeting
Recommend condition.The technology can be realized flexible information batch and push, and compared to the point-to-point recommended technology of tradition, have in efficiency big
Width is promoted.
Application No. is the Chinese invention patent applications of 200880111967.X " recommendation based on media " to disclose reception table
Up to user to the information of the interest of one or more media programs；Obtain instruction in response to received information multiple media sections
Information of the mesh in the popularity in a human world for being different from the user；And from received information-related the multiple matchmaker
The one or more recommendations to media program of body segment mesh transmission are for being shown to the user.
Application No. is 200910188600 Chinese invention patent application " a kind of Online Video recommended method and video portals
In service system ", the described method comprises the following steps: user terminal sends on-demand request to video portal server；Video door
Family server dynamically calculates the weight of each visual classification according to the passing historical record of the user terminal；Video portal clothes
Business device carries out video recommendations by the height of classified weight.The system comprises: terminal and User Manager；Video resource is carried out
Storage management and the resource manager of scoring；One storage video recommendations policing policy library；Store the ratings data media of video
Rating database, and the history viewing record customer data base of storage user；The policy library presses scheduled video recommendations plan
Slightly recommend video to user.The Online Video recommended technology that the invention provides, algorithm are not static but dynamic, systems
In data with user's amount of access increase dynamically self adjustment and improvement, by splendid user experience, to improve use
The stickiness at family.
However, though information recommendation technology comparison in the prior art is more, be mainly used in based on user's purchasing power and
The product advertising push of demand, news or audio/video multimedia commending contents based on browsing viewing history and user interest pass
Broadcast, the analysis of group based on social network relationships with recommend etc., lack the course suitable for Network Education System and recommend
Correlation technique and system.
For Network Education System, it is desirable to be able to for learning objective, the pedagogic objective, knowledge of each student's individual
Situation and personal interest are grasped in basis, lessons and study wish realizes that target, efficient course are recommended, therefore there is an urgent need to one
The method and its realize system that the special recommendation information of kind generates.
Summary of the invention
In view of above-mentioned problem above existing in the prior art, the present invention proposes a kind of based on student's history and real-time learning
State parameter realizes the method and system that course is recommended.Course of the present invention first against mass data grade in E-learning Platform
Resource establishes calculation of relationship degree method；And then merge the study history and learning state statistical number of magnanimity student on entire platform
According to the standard student portrait model and its correlated curriculum set of generation；For as course recommend target object student, according to
Its learned lesson history, statistics establish study history histogram；For real-time for the student for recommending target object as course
Learning behavior parameter, learning Content parameter and study performance parameter, statistics in habit state establish learning state histogram, the two
Simultaneous forms student's feature histogram；With sorting algorithm, target object and standard student portrait model is recommended to carry out course
The classification of histogram distribution feature is compared, and determines that course recommends the standard student that is belonged to of target object to draw a portrait model, and
Standard student draws a portrait on the basis of the corresponding correlated curriculum set of model, is determined according to the degree of association to current target object student's
Course is recommended.
The present invention proposes a kind of method for realizing that course is recommended based on student's history and real-time learning state parameter, feature
It is, comprising:
Step 1, for the course resources of mass data grade in E-learning Platform, description mark is generated for each course resources
Remember information, descriptive markup information is the label for indicating that course resources are described in multiple dimensions；According to scheduled course
Course resources are organized into using course resources tree as the data of basic unit by the descriptive markup information of learning planning and course resources
Structure, each course resources are as the node in course resources tree；Node based on each course resources of expression is in course resources
Position in tree determines the degree of association parameter between course resources；
Step 2, original standard student portrait model is predefined, each original standard student portrait model indicates to exist
Carry out learning and being in the imaginary student of a certain attainment level, original standard student portrait model in some major field
It include: the study history quantized data and learning state quantized data being distributed on each node of course Planning Standard tree；Root
According to the individual study history quantized data and individual learning state quantized data of magnanimity student, the original standard student is drawn
As model is modified；For the standard student portrait model corrected, is formed and standard student portrait model is corresponding related
Course set；
Step 3, for the individual student for recommending target object as course, historical data is learnt according to it, statistics is established
Learn history quantized data histogram；For the study performance parameter for indicating that course recommends target object real-time learning state, system
Meter establishes learning state quantized data histogram, and the two simultaneous forms student's feature histogram；With sorting algorithm, course is pushed away
Target object is recommended compared with the classification of standard student portrait model progress histogram distribution feature, determines that course recommends target object
The standard student portrait model belonged to；
Step 4, the corresponding correlated curriculum set of standard student portrait model for recommending target object to be belonged in course
On the basis of, determine that the course to current target object student is recommended according to the degree of association.
Preferably, in step 1, the descriptive markup information includes: Resource Properties descriptive markup, description of resource content
Label, resource additional information descriptive markup, resource grading descriptive markup.
Preferably, in step 1, the data structure for defining course Planning Standard tree is advised Course Category as course
Draw the node of standard tree；For each node of course Planning Standard tree, corresponding course resources search condition is generated；It utilizes
The course resources search condition is retrieved in each course resources description of resource content label generated, will be examined
Rope to course resources belong to the node of course Planning Standard tree；By each node for belonging to course Planning Standard tree
Course resources form tree data structure, as the course resources tree.
Preferably, in step 1, for any two course resources, to indicate this two on the course resources tree
The mutual nodal distance of the node of course resources calculates between two course resources in conjunction with the similarity of course resources
Degree of association parameter；Degree of association parameter between two course resources is with the nodal distance and the phase of the course resources
It is indicated like the two-dimensional array that degree is constituted.
Preferably, in step 2, the study history quantized data of standard student portrait model includes being distributed in often
Study duration or learned lesson number on a node, learning state quantized data include the study performance of distribution on each node
Parameter.
Preferably, in step 2, original standard student draws a portrait model with predefined study history quantized data
And learning state quantized data；And it is individual for the mass users for some standard student portrait model for being determined ownership,
Each user's individual is extracted to learn in the individual that standard student draws a portrait on each node of course Planning Standard tree that model includes
History quantized data and individual learning state quantized data are practised, and then calculates the course rule for including in standard student portrait model
Draw each node of standard tree on the study history quantized data of mass users individual and the mean value of learning state quantized data, and according to
Original standard student portrait model is modified according to the mean value.
Preferably, in step 3, with svm classifier algorithm, course is recommended into target object and standard student portrait model
The classification for carrying out histogram distribution feature is compared, and determines the standard student portrait model that course recommends target object to be belonged to.
Preferably, in step 4, the course resources for recommending user's individual of target object to learn as course are obtained,
At least part is therefrom chosen as recommendation and refers to course；It determines in the standard student portrait model belonged to user individual
In relevant course set, course resources of the nodal distance of course within predetermined distance threshold are referred to recommendation, as
Recommend syllabus；Within the scope of the course resources for recommending syllabus to be included, determine within the scope of this each course resources with
Recommend the sequencing of similarity with reference to course, and is generated according to sequencing of similarity and recommend curriculums table.
Preferably, when determining to recommend syllabus, the junior direction for recommending the node with reference to where course will be only located at
On course resources within predetermined distance threshold of node and nodal distance as recommending syllabus.
The present invention also provides a kind of information recommendation systems, for being realized based on student's history and real-time learning state parameter
Course is recommended characterized by comprising
Course resources data server, the course resources file for including for storing E-learning Platform；
Course resources label and Analysis server, each course resources for including for E-learning Platform generate description
Mark information, including Resource Properties descriptive markup, description of resource content label, the grading of resource additional information descriptive markup, resource
Descriptive markup is saved as each course resources in the form of database file and generates descriptive markup information；Also, by course resources
Be organized into using course resources tree as the data structure of basic unit, each course resources as the node in course resources tree, with
The form of database index file saves the data structure of course resources tree；Node based on each course resources of expression is in course
Position in resource tree determines incidence relation and its degree of association parameter between course resources；
Information of trainee management server, the learning records information that course resources are learnt for storing user's individual；
Information of trainee Analysis server, for establishing and correcting standard student portrait model, which includes being distributed in class
Study history quantized data and learning state quantized data on each node of journey Planning Standard tree；For the standard corrected
Student's portrait model forms standard student portrait mould in conjunction with the course resources tree that course resources label is obtained with Analysis server
Type correlated curriculum set；Learning records information is called from information of trainee management server, calculates and recommends target object as course
User's individual individual study history quantized data and individual learning state quantized data；According to the study history of user's individual
Quantized data and individual learning state quantized data, statistics establish study history quantized data histogram on each node and
Learn performance quantization data graphs, histogram distribution feature of the two simultaneous as user's individual；It is drawn a portrait according to standard student
The histogram distribution feature of the histogram distribution feature of model and user's individual is determined as the user that course recommends target object
The standard student portrait model that individual is belonged to；
Recommendation information generates and push server, obtains from the information of trainee Analysis server and recommends target as course
The standard student portrait model that user's individual of object is belonged to, and in the corresponding correlated curriculum of standard student portrait model
On the basis of set, is determined according to the degree of association and recommend the course of target object to recommend to current course；Generate corresponding recommendation
It ceases and is transferred to the terminal for recommending user's individual of target object as course.
As it can be seen that the present invention provides it is a kind of specialization suitable for E-learning Platform course resources recommended method and be
System, from general or other fields information recommendation method and system it is different, the present invention is to the sea on E-learning Platform
Amount course resources form tree-shaped data structure, and the school work feature for comprehensively considering professional knowledge system and magnanimity student is formed
User draws a portrait model, realizes that classification mode identifies using the study history and learning state of individual student on this basis, into
And the degree of association of the classification according to belonging to student and course, personalized course is generated in big data course resources library to be pushed away
It recommends, to improve the efficiency that student carries out autonomous course selection, saves and determine the current stage from vast as the open sea course material
The time of learning Content, enhance course and student matching degree, so that study schedule is imbued with systemic property and science, finally realize because
Material is taught the basic concept of this education.
Figure of description
Fig. 1 is the overall flow figure of course recommended method of the present invention；
Fig. 2 is the calculation method flow diagram that the course resources degree of association is determined in course recommended method of the present invention；
Fig. 3 is course resources data tree structure schematic diagram；
Fig. 4 is the stream that standard student portrait model and its correlated curriculum set are determined in course recommended method of the present invention
Journey schematic diagram；
Fig. 5 is the data structure schematic diagram for indicating standard student portrait model；
Fig. 6 is the mark for determining course in course recommended method of the present invention and the individual student of target object being recommended to be belonged to
The flow chart of quasi- student's portrait model；
Fig. 7 A-7B be individual student individual study history quantized data histogram, individual study performance quantization data it is straight
The schematic diagram of square figure；
Fig. 8 A-8B be standard student draw a portrait model study history quantized data histogram, study performance quantization data it is straight
The schematic diagram of square figure；
Fig. 9 is in course recommended method of the present invention in the base of the corresponding correlated curriculum set of standard student portrait model
The flow chart that course is recommended is determined on plinth；
Figure 10 is the structural schematic diagram of information recommendation system of the present invention.
Specific embodiment
Below by embodiment, technical solution of the present invention is described in further detail.
The purpose of the present invention is in the big data course resources library of E-learning Platform, towards each student's individual
(i.e. course recommendation target object) generates personalized course and recommends.On the one hand the course recommended can completely cover specific special
On the other hand basic knowledge system necessary to the study in industry field meets student based on study history and is formed by school work basis
Level, and match with learning behavior, learning Content and the study performance within the current certain phase of student, it can be with student
The desirability of study and subjective tendency are sufficiently coordinated, and keep the persistent and gradual of learned lesson, avoid generation zero
The recommendation for dissipating fragmentation, avoids the recommendation for being detached from student's self-condition from course content merely, avoids learning away from student
Practise target, direction and recommendation of interest etc..
The present invention provides a kind of methods for realizing course recommendation based on student's history and real-time learning state parameter, such as scheme
Overall flow shown in 1, in which:
Firstly, for the course resources of mass data grade in E-learning Platform, course resources are organized into step 1
Using course resources tree as the data structure of basic unit；Based on calculation of relationship degree method, the mutual pass between course resources is calculated
Connection degree；
In turn, in step 2, the study historical data and learning state data of magnanimity student on entire platform are merged, is generated
Multi-direction, multiple stratification standard student portrait model, and correlated curriculum set is formed for each standard student portrait model；
In step 3, for the individual student for recommending target object as course, historical data is learnt according to it, statistics is built
Vertical study history quantized data histogram；For the study achievement of the real-time learning state for the student for recommending target object as course
Parameter is imitated, statistics establishes learning state quantized data histogram, and the two simultaneous forms student's feature histogram；It is calculated with classification
Method determines course compared with recommending target object to carry out the classification of histogram distribution feature with standard student portrait model course
The standard student portrait model for recommending target object to be belonged to.
In step 4, on the basis of standard student portrait model corresponding correlated curriculum set, given according to degree of association determination
The course of current target object student is recommended.
Each step of the method for the present invention is illustrated below.
Fig. 2 is the calculation method stream that course recommended method of the present invention determines the course resources degree of association in step 1
Journey schematic diagram.
In a step 101, descriptive markup information is generated for each course resources first, descriptive markup information is indicated more
The label that course resources are described in a dimension, specifically includes: Resource Properties descriptive markup, description of resource content label,
Resource additional information descriptive markup, resource grading descriptive markup.Resource Properties descriptive markup indicates course resources in file format
(such as the multimedia files such as the text files such as video, audio, PPT or WORD, FLASH, learning software application program etc.), when
Between and length (such as video, the overall duration of audio or multimedia file, the number of pages of text file, the data volume of application program
Size etc.), the attribute in terms of file play quality (such as video resolution, audio compression rate).Description of resource content marks generation
Subject classification, Course in English, main contents and the recognition learning phase for the course set that table course resources are included, for example, course
Resource " lecture of intellectual property law outline " description of resource content label may include: describe subject sort out label " science of law ",
" civil and commercial law ", " intellectual property law " describes the label " intellectual property law outline " of Course in English, describes the label of main contents
" trademark law rudimentary knowledge ", " Patent Law basic knowledge ", " Authorship Right Law basic knowledge " describe the label of recognition learning phase " just
Grade ", " basis ", " introduction "；The description of resource content label of course resources " Patent Law introduction " may include: that description subject is sorted out
Label " science of law ", " Civil Law and Commercial Law ", " intellectual property law ", " Patent Law ", the label " Patent Law introduction " of Course in English is described,
Label " Patent Law general provisions ", " patent application and the examination process ", " license material condition ", " patent of main contents are described
Infringement and lawsuit ", " patent related international treaties ", describe the label " middle rank " of recognition learning phase；Course resources " patent examination essence
Say ", the label " science of law " of description subject classification, " civil and commercial law ", " intellectual property law ", " Patent Law " describes the mark of Course in English
Remember " patent examination is presented briefly and succinctly ", describes label " examination as to novelty ", " creativeness examines ", " practicability examination " of main contents, retouch
State label " advanced ", " advanced ", " special topic " of recognition learning phase；Course resources " patent agency practice strategy ", description subject are sorted out
Label " science of law ", " civil and commercial law ", " intellectual property law ", " Patent Law " describes the label " patent agency practice " of Course in English,
Label " update search method ", " application documents write skill ", " the examining that notice is replied and modified " for describing main contents, retouch
State label " advanced ", " advanced ", " special topic " of recognition learning phase.Resource additional information descriptive markup includes: to indicate that course material is made
Person, editor mechanism, developer or speaker's name label, indicate that course material is applicable in grade and label (such as undergraduate course of crowd
Primary grades, undergraduate course senior class, master, doctor, catechumen, general practitioners, intermediate engineer, senior engineer etc.).Resource is commented
Grade descriptive markup can be by acquiring and completing the user of this course resources viewing on statistics network teaching platform to the resource
Artificial grading to obtain, grading can be related to the difficulty of learning degree evaluation (difficult, in, easily) of the course resources, curriculum quality is commented
Valence (favorable comment, common, difference are commented), recommendation evaluation (recommend very much, general recommendation, do not recommend).Network Education System can be class
The publisher of Cheng Ziyuan or manager provide the index page of criteria table form, to realize the index to course resources,
Generate the descriptive markup information of each dimension.
In a step 102, according to the descriptive markup information of scheduled course learning planning and course resources, by course resources
It is organized into using course resources tree as the data structure of basic unit, each course resources are as the node in course resources tree.Class
Journey resource tree is the data structure for indicating the course resources in big data resources bank and associating with one another.For each profession neck
Relationship between the predetermined Course Category and each Course Category for needing to cover of course planning in domain, defines course Planning Standard tree
Data structure characterized, wherein node of each Course Category as course Planning Standard tree.For example, for intellectual property
Profession, course Planning Standard tree is as shown in figure 3, " intellectual property basic knowledge course " this Course Category plans mark as course
The root node N of quasi- tree, by " patent basic knowledge course ", " trade mark basic knowledge course ", " copyright basic knowledge course " etc.
Course Category is as level-one child node N1, N2, N3, by " patent deep knowledge course ", " trade mark deep knowledge course ", " works
The Course Categories such as power deep knowledge course " are as second level child node N11, N21, N31.Course Planning Standard tree is that Web education is flat
Platform data structure predetermined according to the Course Exercise of professional domain.For each node of course Planning Standard tree,
Generate corresponding course resources search condition；For example, for " intellectual property basic knowledge course " as root node, Ke Yisheng
At keyword " intellectual property " as represent subject sort out corresponding with Course in English description of resource content mark in retrieve
Condition, generate keyword " primary ", " basis " is retrieved as in the description of resource content for representing recognition learning phase marks
Condition；Similar, for " the patent deep knowledge course " as second level child node, keyword " patent " conduct can be generated
Sort out the condition retrieved in description of resource content label corresponding with Course in English representing subject, generates keyword " advanced "
As the condition retrieved in the description of resource content label for representing recognition learning phase.Using in course Planning Standard tree
The corresponding course resources search condition of each node, in the resource generated for each course resources in big data resources bank
Hold and retrieved in descriptive markup, the course resources retrieved are belonged to the node of course Planning Standard tree；To, by
The course resources for belonging to each node of course Planning Standard tree form tree data structure, as course resources tree, such as Fig. 3
It is shown.By this step, the course resources after label are organized into using course resources tree as the data structure of basic unit.
In step 103, based on position of the node of each course resources in course resources tree is indicated, determine that course provides
Incidence relation and its degree of association parameter between source.This method will indicate the mutual distance knot of each node of course resources
Close the degree of association parameter between the similarity calculation course resources of course resources.As shown in figure 3, in course resources tree, class
Cheng Ziyuan " lecture of intellectual property law outline " belongs to root node N, and course resources " Patent Law introduction " belong to level-one child node
N1, and course resources " patent examination is presented briefly and succinctly " and course resources " patent agency practice strategy " belong to the same second level child node
N11 can determine course resources " lecture of intellectual property law outline " and course according to distance of the node in tree data structure
The distance between resource " Patent Law introduction " is that (k is the fundamental distance in tree between two adjacent the superior and the subordinate's nodes to k
Linear module), then course resources " patent examination is presented briefly and succinctly " and course resources " patent agency practice strategy " and course resources " patent
The distance between method introduction " is k, course resources " patent examination is presented briefly and succinctly " and course resources " patent agency practice strategy " and course
The distance between resource " lecture of intellectual property law outline " is 2k, and the distance between the above node is as the association between course resources
Spend parameter.It, can be according to Resource Properties descriptive markup, the resource for describing course resources meanwhile for any two course resources
Content descriptive markup, resource additional information descriptive markup and resource grading descriptive markup, utilize weighting algorithm to calculate course and provide
The similarity in source, and similarity is also used as to the degree of association parameter between two course resources.It specifically, can be according to as follows
Formula calculates the similarity between two course resources:
A=αP·DP+αC·DC+αA·DA+αR·DR
Wherein, DPSimilar scoring of any two course resources in terms of Resource Properties is indicated, for example, if two courses
The Resource Properties descriptive markup of resource belongs to video, then the similar scoring is highest (such as 100), if one belongs to video one
It is a to belong to audio, then similar scoring lower (such as 60), if one belongs to video, audio or multimedia file and another belongs to
Text file, then similar scoring minimum (such as 20)；Similarly, similar if two course resources file play qualities are close
Scoring is high, and on the contrary then similar scoring is low；The mapping table of comparisons can be established to save the above Rule of judgment and corresponding similar scoring；αP
Then indicate weighting weighted value of two course resources in the similar scoring in terms of Resource Properties when calculating similarity.DCIt indicates to appoint
Similar scoring of two course resources of anticipating in terms of resource content, marks, the keyword of judge mark according to description of resource content
Repetitive rate, the more high then scoring of repetitive rate are higher；αCThen indicate that similar scoring of two course resources in terms of resource content exists
Calculate weighting weighted value when similarity.DAIndicate similar scoring of any two course resources in terms of resource additional information,
Analogously, course resources from identical course material author, editor mechanism, developer or speaker's name or towards phase
Same grade/crowd course resources scoring height, the scoring is low if different；αAThen indicate that two course resources are attached in resource
Add weighting weighted value of the similar scoring of message context when calculating similarity.DRIndicate that any two course resources are commented in resource
The similar scoring of grade aspect, the grading of the same category is closer, then the scoring is high, and the scoring is low if the grading difference the big, αR
Then indicate weighting weighted value of two course resources in the similar scoring in terms of resource grading when calculating similarity.
On the basis of above calculate, the degree of association parameter between any two course resources can use the course resources
It is indicated in the two-dimensional array that the nodal distance in course resources tree and the similarity A between course resources are constituted.
Fig. 4 is multi-direction, multiple stratification standard student portrait model to be generated in step 2, and draw for each standard student
As model forms the flow diagram of correlated curriculum set.Firstly, in step 201, Network Education System can predefine original
Standard student draw a portrait model, each original standard student model expression of drawing a portrait learns in some major field
And it is in the imaginary student of a certain attainment level.Original standard student portrait model includes: to be distributed in course Planning Standard
Study history quantized data and learning state quantized data on the destined node of tree；Study history quantized data includes being distributed in
Study duration or learned lesson number on each node, learning state quantized data include the study achievement of distribution on each node
Imitate parameter.For example, as shown in figure 5, learn and be in patent direction the imaginary student institute of introduction attainment level
The original standard student established draws a portrait in model, and study history quantized data includes the root section for being distributed in course Planning Standard tree
Point N, level-one child node N1, the learned lesson number on second level child node N11, i.e., what student had learnt is subordinated to the class of the node
The quantity of Cheng Ziyuan is such as 8 in the learned lesson number of root node N, and the learned lesson number of level-one child node N1 is 4, second level sub- section
The learned lesson number of point N11 is 0；Learning state quantized data includes the study performance parameter being distributed on above-mentioned each node,
The corresponding test test result mean value obtained of the course resources that performance participates in the subordinate node with student indicates, such as in root node
The study performance parameter of N is 60, is 40 in the study performance parameter of level-one child node N1, in the study achievement of second level child node N11
Imitating parameter is 0.Similar, it is established in the imaginary student that patent direction learn and in advanced attainment level original
Standard student draw a portrait in model, the learned lesson number of root node N is 2, and the learned lesson number of level-one child node N1 is 6, second level
The learned lesson number of child node N11 is 8；The study performance parameter of root node N is 90, is joined in the study performance of level-one child node N1
Amount is 80, is respectively 70 in the study performance parameter of second level child node N11.
In step 202, learnt according to the individual study history quantized data of magnanimity student on E-learning Platform and individual
State quantized data is modified the original standard student portrait model, to generate the standard of constantly dynamic adjustment
Student's portrait model.Student practical for either one or two of E-learning Platform quantifies number according to the study history of individual student
According to learning state quantized data, determine individual student learns in which major field and in which grade
Attainment level-determine standard student that the student belonged to draw a portrait that (specific determination method will be carried out hereinafter model
Introduce) after, it extracts individual student and draws a portrait on each node of course Planning Standard tree that model includes in standard student
Learn history quantized data and learning state quantized data.For example, having determined that it belongs to special for a certain individual student Zhang San
Sharp direction learn and the practical student in introduction attainment level, root node N of the student in course Planning Standard tree
Learned lesson number is 12, and the learned lesson number of level-one child node N1 is 6, and the learned lesson number of level-one child node N3 is 1, second level
The learned lesson number of node N11 is 1, and the learned lesson number of second level child node N21 is 0, the learned lesson number of second level child node N31
It is 1, the expression patent direction which is belonged to learns and the standard student portrait mould in introduction attainment level
Course Planning Standard tree node that type is included as it was noted above, be root node N, level-one child node N1, second level child node N11,
Learned lesson number of the Zhang San on root node N, level-one child node N1 and second level child node N11 is then extracted, and ignores it in level-one
The learned lesson number of child node N3 and second level child node N31.It is similar, extract Zhang San in root node N, level-one child node N1 and
Study performance parameter on second level child node N11, the study performance parameter of root node N is 65, in the study of level-one child node N1
Performance parameter is 50, is respectively 40 in the study performance parameter of second level child node N11, ignores it in level-one child node N3 and second level
The study performance parameter of child node N31.For the magnanimity student of platform, can according to for each student extract in standard learning
The study history quantized data and learning state quantization number on each node of course Planning Standard tree that member's portrait model includes
According to, calculate standard student draw a portrait model include each node of course Planning Standard tree on study history quantized data and
Student's mean value of habit state quantized data, and original standard student portrait model is modified according to the mean value, thus raw
At the standard student portrait model of continuous dynamic adjustment.For example, learn and in introduction school work in patent direction through counting
Average learned lesson number of the horizontal magnanimity student on root node N, level-one child node N1 and second level child node N11 is as follows: root
Node N is 10, level-one child node N1 is 7, second level child node N11 is 2；Learn and in introduction school work water in patent direction
Average study performance parameter of the flat magnanimity student on root node N, level-one child node N1 and second level child node N11 is as follows: root
Node N is 62, level-one child node N1 is 53, second level child node N11 is 38.By this be averaged learned lesson number and averagely learn performance
Interval Maps of the parameter according to locating for it are converted to modifying factor, and with modifying factor multiplied by original standard student portrait model
Predefined average learned lesson number and averagely study performance parameter obtain the standard student portrait model of constantly dynamic adjustment.
For example, carrying out the standard student portrait mould learnt and in introduction attainment level in patent direction after correcting with modifying factor
In type, study history quantized data includes the root node N for being distributed in course Planning Standard tree, level-one child node N1, second level sub- section
Learned lesson number and study performance parameter on point N11, wherein the learned lesson number of root node N is modified to 9, level-one sub- section
The learned lesson number of point N1 is modified to 6, and the learned lesson number of second level child node N11 is modified to 1；The study performance of root node N is joined
Amount is modified to 61,48 is modified in the study performance parameter of level-one child node N1, in the study performance parameter of second level child node N11
It is modified to 25.
In step 203, for the standard student portrait model corrected, correlated curriculum set is formed.Hereinbefore, it introduces
The node of course Planning Standard tree has the course resources for belonging to the node, thus, it draws a portrait according to each standard student
Each node of the corresponding course Planning Standard tree of model can determine standard student portrait model correlated curriculum set.Such as
It is described above, learn in patent direction and draws a portrait model corresponding to course planning in the standard student of introduction attainment level
Therefore root node N, level-one child node N1, the second level child node N11 of standard tree can will belong to root node N, level-one sub- section
Point N1, second level child node N11 course resources as patent direction carry out learn and in introduction attainment level standard learning
The relevant course set of member's portrait model.
Fig. 6 is the standard student portrait for determining course in step 3 of the present invention and the individual student of target object being recommended to be belonged to
The flow chart of model.Firstly, in step 301, obtaining individual student (such as Zhang San) by having learned on E-learning Platform
The course resources of habit and the study history quantized data formed on each node of course Planning Standard tree, as described above,
Zhang San is 12 in the learned lesson number of root node N, and the learned lesson number of level-one child node N1 is 6, the study of level-one child node N2
Course number is 0, and the learned lesson number of level-one child node N3 is 1, and the learned lesson number of second level child node N11 is 1, second level child node
The learned lesson number of N21 is 0, and the learned lesson number of second level child node N31 is 1；Learn history quantized data according to it, statistics is built
The study history quantized data histogram on each node is found, as shown in Figure 7 A.In step 302, continue to obtain the individual
Student (such as Zhang San) is and the study performance of the course resources learnt on E-learning Platform in course Planning Standard tree
Each node on the study performance parameter that is formed, such as Zhang San is 65 in the study performance parameter of root node N, in level-one sub- section
The study performance parameter of point N1 is 50, is 0 in the study performance parameter of level-one child node N2, in the study achievement of level-one child node N3
Imitating parameter is 80, is respectively 40,0 and 35 in the study performance parameter of second level child node N11-N31；Statistics is established in each node
On study performance quantization data graphs, such as Fig. 7 B.In step 303, by study history quantized data histogram and study achievement
Both effect quantized data histograms simultaneous gets up, and student's feature histogram of individual student Zhang San is formed, as individual student's
Histogram distribution feature.In step 304, draws a portrait for each standard student that this platform is established and be distributed in class possessed by model
Study history quantized data and learning state quantized data on each node of journey Planning Standard tree count and determine standard student
The histogram distribution feature of portrait model, for example, showing such as Fig. 8 A-8B and learn and in introduction in patent direction
Histogram distribution of the standard student portrait model of industry level on study history quantized data and learning state quantized data is special
Histogram simultaneous of the standard student portrait model on study history quantized data and learning state quantized data is got up to make by sign
For the histogram distribution feature of standard student portrait model.In step 304, with svm classifier algorithm, course is recommended into target pair
Compared with carrying out the classification of histogram distribution feature with standard student portrait model, determine that course recommends target object to be belonged to
Standard student portrait model.Support vector machines (SVM) sorting algorithm is the supporting vector algorithm for solving pattern recognition problem, existing
SVM classifier for realizing sample classification is provided in technology, the histogram distribution feature of standard student portrait model is substituted into
SVM classifier is trained, and obtains Classification and Identification data；Course is recommended to the histogram distribution of the individual student of target object again
Feature is compared with Classification and Identification data, so that the class categories of the histogram distribution feature of individual student are obtained, according to this
Class categories determine its standard student portrait model belonged to.The standard learning belonged to according to the individual student that this step determines
After member's portrait model, can also by the study history quantized data of individual student and learning state quantized data applied to pair
The amendment of standard student portrait model.
In step 4, it on the basis of standard student portrait model corresponding correlated curriculum set, is determined according to the degree of association
Recommend to the course of current target object student.It identifies in step 3 and recommends the individual of target object to learn as current course
The standard student that member is belonged to draws a portrait after model, then can obtain course set relevant to standard student portrait model.
As it was noted above, present invention course resources are in the nodal distance in course resources tree and the similarity A between course resources
The two-dimensional array of composition indicates the degree of association parameter between two course resources.In order to meet the study of particular professional field must
Target object is recommended to recommend specific course resources to course on the basis of the knowledge hierarchy and student's school work foundation level that need, and
And coordinate the subjective tendency within the course resources recommended and the current certain phase of student sufficiently, make student to the class recommended
Cheng Ziyuan keeps interest, thus draws a portrait the relevant course set of model for standard student, by present in the set and with
There are the course resources that student has learnt the course resources of the larger degree of association to generate recommendation list.Specifically, as shown in figure 9,
In step 401, for the course resources that student has learnt, at least part is therefrom chosen as recommendation and refers to course, such as is selected
Course resources completion study in the nearest certain time of student or learnt are taken, the course divided of not up to passing through examination provides
Source, higher course resources of student's grading etc..In step 402, each is recommended to refer to course, determines and is drawn in standard student
It is provided as referring to course of the nodal distance of course within predetermined distance threshold in the relevant course set of model with the recommendation
Source, as recommendation syllabus；For example, course resources " patent examination is presented briefly and succinctly " refer to course, predetermined distance threshold as recommendation
For k, then course resources " Patent Law introduction " are being recommended within syllabus, opposite course resources " lecture of intellectual property law outline "
Because distance is 2k therewith, then do not recommending within syllabus.In addition, considering nodal point separation when determining to recommend syllabus
From while can also further consider node level direction, such as will only be located at the junior side for recommending the node with reference to where course
The upward course resources of node and nodal distance within predetermined distance threshold are as recommendation syllabus；Such as " Patent Law
Introduction " refers to course as recommendation, and predetermined distance threshold k then will only be located at the course resources " patent examination on downstream site
Present briefly and succinctly " and course resources " patent agency practice strategy " conduct recommendation syllabus, without the course that will be located on superior node
Resource " lecture of intellectual property law outline " is as recommendation syllabus.In step 403, in the course for recommending syllabus to be included
It in scope of resource, determines each course resources within the scope of this and the similarity A with reference to course is recommended to sort, will sort highest one
A or several course resources, which generate, recommends curriculums table, is supplied to student and chooses.
The present invention provides a kind of information recommendation system in turn, for real based on student's history and real-time learning state parameter
Existing course is recommended.As shown in Figure 10, which specifically includes: course resources data server 1001, for storing Web education
The course resources file that platform includes, in a distributed manner or based on the medium storage server of centralization, function is the server
It saves as course resources such as the video of course resources, audio, text file, multimedia file, learning software application programs
Data, and the data delivery service that the user (student) to file a request provides downloading or plays online.
Course resources label and Analysis server 1002, each course resources for including for E-learning Platform generate
Descriptive markup information, including Resource Properties descriptive markup, description of resource content label, resource additional information descriptive markup, resource
Grading descriptive markup is saved as each course resources in the form of database file and generates descriptive markup information；Also, by course
Resource is organized into using course resources tree as the data structure of basic unit, and each course resources are as the section in course resources tree
Point saves the data structure of course resources tree in the form of database index file；Based on the node for indicating each course resources
Position in course resources tree determines incidence relation and its degree of association parameter between course resources.
Information of trainee management server 1003 is that the user (i.e. student) of E-learning Platform stores registration information and student
Personal information, the personal information include: the learning records information that user learns course resources, such as the course money of study
Source name, study duration；Rating information of the user to course resources；User's test test obtained corresponding to course resources at
Achievement.
Information of trainee Analysis server 1004, for establishing and correcting standard student portrait model, which includes distribution
In study history quantized data and learning state quantized data on each node of course Planning Standard tree；For what is corrected
Standard student portrait model forms standard learning in conjunction with the course resources tree that course resources label is obtained with Analysis server 1002
Member's portrait model correlated curriculum set；Student's personal information is called from information of trainee management server 1003 and is analyzed, and is counted
It can be regarded as the individual study history quantized data and individual learning state quantized data for recommending user's individual of target object for course；
According to the study history quantized data of user's individual and individual learning state quantized data, statistics establishes on each node
Practise history quantized data histogram and study performance quantization data graphs, histogram distribution of the two simultaneous as user's individual
Feature；The histogram distribution feature of model of being drawn a portrait according to standard student and the histogram distribution feature of user's individual, are determined as
The standard student portrait model that course recommends user's individual of target object to be belonged to；
Recommendation information generates and push server 1005, obtains from the information of trainee Analysis server 1004 as course
The standard student portrait model for recommending user's individual of target object to be belonged to, and it is corresponding in standard student portrait model
On the basis of correlated curriculum set, is determined according to the degree of association and recommend the course of target object to recommend to current course；It generates corresponding
Recommendation information and be transferred to as course recommend target object user's individual terminal.
As it can be seen that the present invention provides it is a kind of specialization suitable for E-learning Platform course resources recommended method and be
System, from general or other fields information recommendation method and system it is different, the present invention is to the sea on E-learning Platform
Amount course resources form tree-shaped data structure, and the school work feature for comprehensively considering professional knowledge system and magnanimity student is formed
User draws a portrait model, realizes that classification mode identifies using the study history and learning state of individual student on this basis, into
And the degree of association of the classification according to belonging to student and course, personalized course is generated in big data course resources library to be pushed away
It recommends, to improve the efficiency that student carries out autonomous course selection, saves and determine the current stage from vast as the open sea course material
The time of learning Content, enhance course and student matching degree, so that study schedule is imbued with systemic property and science, finally realize because
Material is taught the basic concept of this education.
Above embodiments are merely to illustrate the present invention, and not limitation of the present invention, the common skill in relation to technical field
Art personnel can also make a variety of changes and modification without departing from the spirit and scope of the present invention, therefore all etc.
Same technical solution also belongs to scope of the invention, and scope of patent protection of the invention should be defined by the claims.
Claims (8)
1. a kind of method for realizing that course is recommended based on student's history and real-time learning state parameter characterized by comprising
Step 1, for the course resources of mass data grade in E-learning Platform, descriptive markup letter is generated for each course resources
Breath, descriptive markup information is the label for indicating that course resources are described in multiple dimensions, comprising: Resource Properties description mark
Note, description of resource content label, resource additional information descriptive markup, resource grading descriptive markup；According to scheduled course learning
The descriptive markup information of planning and course resources defines the data structure of course Planning Standard tree, using Course Category as course
The node of Planning Standard tree；For each node of course Planning Standard tree, corresponding course resources search condition is generated；Benefit
With the course resources search condition, retrieved in each course resources description of resource content label generated, it will
The course resources retrieved belong to the node of course Planning Standard tree；By each node for belonging to course Planning Standard tree
Course resources form tree data structure, as course resources tree, to course resources are organized into being with course resources tree
The data structure of basic unit, each course resources are as the node in course resources tree；Based on each course resources of expression
Position of the node in course resources tree determines the degree of association parameter between course resources；
Step 2, original standard student portrait model is predefined, each original standard student draws a portrait model expression a certain
Carry out learning and being in the imaginary student of a certain attainment level, original standard student portrait model packet in a major field
It includes: the study history quantized data and learning state quantized data being distributed on the destined node of course Planning Standard tree；According to
The individual study history quantized data and individual learning state quantized data of magnanimity student, draws a portrait to the original standard student
Model is modified；For the standard student portrait model corrected, related class corresponding to standard student portrait model is formed
Cheng Jihe；
Step 3, for the individual student for recommending target object as course, historical data is learnt according to it, statistics establishes study
History quantized data histogram；For the study performance parameter for indicating that course recommends target object real-time learning state, statistics is built
Vertical learning state quantized data histogram, the two simultaneous form student's feature histogram；With sorting algorithm, course is recommended into mesh
Object is marked compared with the classification of standard student portrait model progress histogram distribution feature, determines that course recommends target object to be returned
The standard student portrait model of category；
Step 4, the basis for the corresponding correlated curriculum set of standard student portrait model for recommending target object to be belonged in course
On, determine that the course to current target object student is recommended according to the degree of association.
2. the method according to claim 1 realizing course and recommending, which is characterized in that in step 1, for any two class
Cheng Ziyuan, to indicate the mutual nodal distance of the node of two course resources on the course resources tree, in conjunction with class
The similarity of Cheng Ziyuan calculates the degree of association parameter between two course resources；The degree of association between two course resources
Parameter is indicated with the two-dimensional array that the similarity of the nodal distance and the course resources is constituted.
3. the method according to claim 2 realizing course and recommending, which is characterized in that in step 2, standard student portrait
The study history quantized data of model includes study duration or the learned lesson number being distributed on each node, learning state
Quantized data includes the study performance parameter of distribution on each node.
4. the method according to claim 3 realizing course and recommending, which is characterized in that in step 2, original standard learning
Member's portrait model has predefined study history quantized data and learning state quantized data；And it is directed to and is determined ownership
Some standard student draw a portrait model mass users individual, extract each user's individual standard student draw a portrait model packet
Individual study history quantized data and individual learning state quantized data on each node of course Planning Standard tree contained, in turn
Calculate the study history of the mass users individual on each node of course Planning Standard tree that standard student portrait model includes
The mean value of quantized data and learning state quantized data, and original standard student portrait model is repaired according to the mean value
Just.
5. the method according to claim 4 realizing course and recommending, which is characterized in that in step 3, classify with sVM and calculate
Method determines course compared with recommending target object to carry out the classification of histogram distribution feature with standard student portrait model course
The standard student portrait model for recommending target object to be belonged to.
6. the method according to claim 5 realizing course and recommending, which is characterized in that in step 4, acquisition is pushed away as course
The course resources that user's individual of target object has learnt are recommended, at least part is therefrom chosen as recommendation and refers to course；It determines
It draws a portrait in the relevant course set of model to the standard student that is belonged to of user individual, with the node for recommending to refer to course
Course resources of the distance within predetermined distance threshold, as recommendation syllabus；In the course for recommending syllabus to be included
In scope of resource, determines each course resources within the scope of this and recommend the sequencing of similarity with reference to course, and arranged according to similarity
Sequence, which generates, recommends curriculums table.
7. the method according to claim 6 realizing course and recommending, which is characterized in that when determining to recommend syllabus,
The node on the junior direction for recommending to refer to course place node and nodal distance will be only located within predetermined distance threshold
Course resources are as recommendation syllabus.
8. a kind of information recommendation system, for realizing that course is recommended based on student's history and real-time learning state parameter, feature
It is, comprising:
Course resources data server, the course resources file for including for storing E-learning Platform；
Course resources label and Analysis server, each course resources for including for E-learning Platform generate descriptive markup
Information, including Resource Properties descriptive markup, description of resource content label, the grading description of resource additional information descriptive markup, resource
Label is saved as each course resources in the form of database file and generates descriptive markup information；Also, according to scheduled course
The descriptive markup information of learning planning and course resources, define course Planning Standard tree data structure, using Course Category as
The node of course Planning Standard tree；For each node of course Planning Standard tree, corresponding course resources retrieval item is generated
Part；Using the course resources search condition, examined in each course resources description of resource content label generated
The course resources retrieved are belonged to the node of course Planning Standard tree by rope；By belonging to each of course Planning Standard tree
The course resources of a node are formed tree data structure and are provided as course resources tree to being organized into course resources with course
Source tree is the data structure of basic unit, and each course resources are as the node in course resources tree, with database index file
Form save course resources tree data structure；Position of the node based on each course resources of expression in course resources tree
It sets, determines the incidence relation and its degree of association parameter between course resources；
Information of trainee management server, the learning records information that course resources are learnt for storing user's individual；
Information of trainee Analysis server, for establishing and correcting standard student portrait model, which includes being distributed in course rule
Draw the study history quantized data and learning state quantized data on each node of standard tree；For the standard student corrected
Portrait model forms standard student portrait model phase in conjunction with the course resources tree that course resources label is obtained with Analysis server
Close course set；Learning records information is called from information of trainee management server, calculates the use for recommending target object as course
The individual study history quantized data and individual learning state quantized data of family individual；Quantified according to the study history of user's individual
Data and individual learning state quantized data, statistics establish study history quantized data histogram and study on each node
Performance quantization data graphs, histogram distribution feature of the two simultaneous as user's individual；According to standard student portrait model
Histogram distribution feature and user's individual histogram distribution feature, be determined as course recommend target object user individual
The standard student portrait model belonged to；
Recommendation information generates and push server, obtains from the information of trainee Analysis server and recommends target object as course
The standard student that is belonged to of user's individual draw a portrait model, and draw a portrait the corresponding correlated curriculum set of model in standard student
On the basis of, it is determined according to the degree of association and recommends the course of target object to recommend to current course；Generate corresponding recommendation information simultaneously
It is transferred to the terminal for recommending user's individual of target object as course.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610911525.XA CN106528656B (en) | 2016-10-20 | 2016-10-20 | A kind of method and system for realizing that course is recommended based on student's history and real-time learning state parameter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610911525.XA CN106528656B (en) | 2016-10-20 | 2016-10-20 | A kind of method and system for realizing that course is recommended based on student's history and real-time learning state parameter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106528656A CN106528656A (en) | 2017-03-22 |
| CN106528656B true CN106528656B (en) | 2019-08-20 |
Family
ID=58332541
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610911525.XA Expired - Fee Related CN106528656B (en) | 2016-10-20 | 2016-10-20 | A kind of method and system for realizing that course is recommended based on student's history and real-time learning state parameter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106528656B (en) |
Families Citing this family (53)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107240046A (en) * | 2017-04-28 | 2017-10-10 | 深圳前海易维教育科技有限公司 | A kind of Learning behavior analyzing method and system |
| CN109241393A (en) * | 2017-07-07 | 2019-01-18 | Ai平台有限公司 | Learning Content based on artificial intelligence machine study recommends automated system |
| CN107358828A (en) * | 2017-09-06 | 2017-11-17 | 华中师范大学 | A kind of classroom interaction tutoring system and method |
| CN107578292B (en) * | 2017-09-19 | 2020-10-16 | 上海财经大学 | User portrait construction system |
| CN107610013A (en) * | 2017-09-23 | 2018-01-19 | 武汉东大思源教育科技有限公司 | A kind of adult education is from test system |
| CN108198474A (en) * | 2018-01-30 | 2018-06-22 | 广东小天才科技有限公司 | A kind of learning data generation method, device and equipment |
| CN108765213A (en) * | 2018-04-20 | 2018-11-06 | 上海学民网络科技有限公司 | A kind of knowledge modularization architecture system and processing method |
| CN108718340A (en) * | 2018-05-24 | 2018-10-30 | 郑州升达经贸管理学院 | A kind of foreign language Digital Learning system and its learning method |
| CN109033438B (en) * | 2018-08-15 | 2022-06-21 | 邢鲁华 | Method and device for recording learning content of user |
| CN109146174A (en) * | 2018-08-21 | 2019-01-04 | 广东恒电信息科技股份有限公司 | A kind of elective course accurate recommendation method based on result prediction |
| CN109271586A (en) * | 2018-08-30 | 2019-01-25 | 上海学民网络科技有限公司 | A kind of knowledge modularization search system and processing method |
| CN109191960A (en) * | 2018-09-25 | 2019-01-11 | 北京盒子鱼教育科技有限公司 | A kind of Teaching Service System and method |
| CN109584976A (en) * | 2018-10-30 | 2019-04-05 | 嘉兴太美医疗科技有限公司 | Drug Warning System and method comprising user's portrait grade evaluation and grade training |
| CN109740861A (en) * | 2018-12-12 | 2019-05-10 | 中国联合网络通信集团有限公司 | A kind of learning data analysis method and device |
| CN109739896B (en) * | 2018-12-12 | 2021-02-02 | 中国联合网络通信集团有限公司 | Data analysis method and device for online learning platform |
| CN110060538A (en) * | 2019-04-08 | 2019-07-26 | 上海云之驾科技股份有限公司 | Personalized artificial based on historical data modeling intelligently drives training and practices system and method |
| CN111950574A (en) * | 2019-05-15 | 2020-11-17 | 北京荣森利泰科贸有限公司 | Ritual training information pushing method and device based on wearable equipment |
| CN110163723A (en) * | 2019-05-20 | 2019-08-23 | 深圳市和讯华谷信息技术有限公司 | Recommended method, device, computer equipment and storage medium based on product feature |
| CN110188279B (en) * | 2019-05-31 | 2021-06-29 | 苏州百智通信息技术有限公司 | Learning resource recommendation method and device |
| CN110442670B (en) * | 2019-06-11 | 2023-05-26 | 天津交通职业学院 | Consumer portrait generation method based on text indexing |
| CN110570699B (en) * | 2019-08-27 | 2021-11-02 | 格局商学教育科技（深圳）有限公司 | Intelligent client management system and method based on live broadcast teaching |
| CN111009311A (en) * | 2019-11-05 | 2020-04-14 | 泰康保险集团股份有限公司 | Medical resource recommendation method, device, medium and equipment |
| CN110990691A (en) * | 2019-11-14 | 2020-04-10 | 泰康保险集团股份有限公司 | Online course recommendation method and device and computer storage medium |
| CN111222792B (en) * | 2020-01-08 | 2022-04-05 | 中国建设银行股份有限公司 | Method and device for realizing hybrid training |
| CN111460249B (en) * | 2020-02-24 | 2022-09-09 | 桂林电子科技大学 | Personalized learning resource recommendation method based on learner preference modeling |
| CN111178790B (en) * | 2020-02-28 | 2023-06-20 | 华兰生物工程重庆有限公司 | Electric power detecting system |
| CN112287228A (en) * | 2020-03-20 | 2021-01-29 | 张明 | Online learning recommendation method, online learning system and server |
| CN111402098B (en) * | 2020-04-20 | 2023-02-28 | 深圳市火火兔智慧科技有限公司 | Intelligent early education method, system, equipment and storage medium based on child growth period |
| CN112765374A (en) * | 2020-07-27 | 2021-05-07 | 上海斐杰教育科技有限公司 | Education resource screening system and method for information push |
| CN111932414A (en) * | 2020-08-07 | 2020-11-13 | 泰康保险集团股份有限公司 | Training management system and method, computer storage medium and electronic equipment |
| CN112102123B (en) * | 2020-08-31 | 2022-09-13 | 武汉美和易思数字科技有限公司 | Student knowledge plane testing method and device based on data matching |
| CN112001656A (en) * | 2020-09-01 | 2020-11-27 | 北京弘远博学科技有限公司 | Method for carrying out training course recommendation pertinently based on employee historical training information |
| CN112085099B (en) * | 2020-09-09 | 2022-05-17 | 西南大学 | Distributed student clustering integration method and system |
| CN112199594A (en) * | 2020-10-14 | 2021-01-08 | 绿瘦健康产业集团有限公司 | Training content pushing method and system |
| CN112307346B (en) * | 2020-11-09 | 2021-06-11 | 深圳市思达优悦企业管理顾问有限公司 | Courseware management system based on big data |
| CN112465678A (en) * | 2020-12-07 | 2021-03-09 | 上海光数信息科技有限公司 | Student course selection recommendation method and system |
| CN112541845B (en) * | 2020-12-15 | 2023-12-01 | 泰康保险集团股份有限公司 | Online training method and device, electronic equipment and storage medium |
| CN112614032B (en) * | 2020-12-28 | 2021-07-20 | 深圳市诺达教育股份有限公司 | Education software data processing method based on big data |
| CN112579909A (en) * | 2020-12-28 | 2021-03-30 | 北京百度网讯科技有限公司 | Object recommendation method and device, computer equipment and medium |
| CN112632393A (en) * | 2020-12-30 | 2021-04-09 | 北京博海迪信息科技有限公司 | Course recommendation method and device and electronic equipment |
| CN112749340A (en) * | 2021-01-19 | 2021-05-04 | 敖客星云(北京)科技发展有限公司 | Course recommendation method and system based on big data, mobile terminal and storage medium |
| CN113077147B (en) * | 2021-03-31 | 2022-06-24 | 国家电网有限公司 | Intelligent student course pushing method and system and terminal equipment |
| CN113077142B (en) * | 2021-03-31 | 2022-12-27 | 国家电网有限公司 | Intelligent student portrait drawing method and system and terminal equipment |
| CN113256464B (en) * | 2021-05-28 | 2022-03-04 | 智慧校园(广东)教育科技有限公司 | Online education course automatic distribution system and method based on cloud computing |
| CN113377925B (en) * | 2021-06-28 | 2022-10-28 | 哈尔滨工业大学 | Online course portrait construction method based on feature engineering |
| CN113762801B (en) * | 2021-09-17 | 2024-03-26 | 北京量子之歌科技有限公司 | Network course management method, device, equipment and storage medium |
| CN114154839A (en) * | 2021-11-29 | 2022-03-08 | 华中师范大学 | Course recommendation method based on online education platform data |
| CN114357411B (en) * | 2022-01-19 | 2022-09-02 | 英才(广州)在线教育科技有限公司 | Online education system based on block chain |
| CN114418807B (en) * | 2022-03-30 | 2022-06-28 | 北京英华在线科技有限公司 | Course recommendation method and system of online education platform based on historical score |
| CN115511677A (en) * | 2022-10-18 | 2022-12-23 | 天津景润网络科技有限公司 | Intelligent education resource platform allocation method and system |
| CN116186705A (en) * | 2022-11-17 | 2023-05-30 | 北京东方通科技股份有限公司 | Software security code analyzer based on static analysis of source code and detection method thereof |
| CN116561421A (en) * | 2023-05-11 | 2023-08-08 | 广东工贸职业技术学院 | Student course recommendation method, device and equipment based on face recognition |
| CN117151346B (en) * | 2023-10-30 | 2024-02-09 | 中国民航大学 | Civil aviation specialty teaching training system based on wisdom study |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1643555A (en) * | 2002-02-11 | 2005-07-20 | Sap股份公司 | Navigation e-course |
| CN101105853A (en) * | 2007-08-16 | 2008-01-16 | 上海交通大学 | Personalized teaching-guiding system based on non-zero jumping-off point in network teaching |
| CN102822882A (en) * | 2010-01-15 | 2012-12-12 | 阿波洛集团公司 | Dynamically recommending learning content |
| CN103886054A (en) * | 2014-03-13 | 2014-06-25 | 中国科学院自动化研究所 | Personalization recommendation system and method of network teaching resources |
| CN105912604A (en) * | 2016-04-05 | 2016-08-31 | 苏州奇展信息科技有限公司 | On-line training platform for customized recommendation of curriculums |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120329030A1 (en) * | 2010-01-29 | 2012-12-27 | Dan Joseph Leininger | System and method of knowledge assessment |
-
2016
- 2016-10-20 CN CN201610911525.XA patent/CN106528656B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1643555A (en) * | 2002-02-11 | 2005-07-20 | Sap股份公司 | Navigation e-course |
| CN101105853A (en) * | 2007-08-16 | 2008-01-16 | 上海交通大学 | Personalized teaching-guiding system based on non-zero jumping-off point in network teaching |
| CN102822882A (en) * | 2010-01-15 | 2012-12-12 | 阿波洛集团公司 | Dynamically recommending learning content |
| CN103886054A (en) * | 2014-03-13 | 2014-06-25 | 中国科学院自动化研究所 | Personalization recommendation system and method of network teaching resources |
| CN105912604A (en) * | 2016-04-05 | 2016-08-31 | 苏州奇展信息科技有限公司 | On-line training platform for customized recommendation of curriculums |
Non-Patent Citations (1)
| Title |
|---|
| 基于属性相关的智能学习指导模型的设计与实现;张春飞 等;《河北科技大学学报》;20130114;第33卷(第6期);525-529 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106528656A (en) | 2017-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106528656B (en) | A kind of method and system for realizing that course is recommended based on student's history and real-time learning state parameter | |
| Thanuskodi | Use of E-resources by the Students and Researchers of Faculty of Arts, Annamalai University | |
| Mitrofanova et al. | Modeling smart learning processes based on educational data mining tools | |
| Jalali et al. | Analytical assessment process of e-learning domain research between 1980 and 2014 | |
| Gobster et al. | Landscape journal and scholarship in landscape architecture: The next 25 years | |
| CN108460139A (en) | Based on web crawlers data mining online course Management System for Evaluation Teaching Quality | |
| Hsieh | Online learning era: Exploring the most decisive determinants of MOOCs in Taiwanese higher education | |
| Pettit et al. | The Australian urban intelligence network supporting smart cities | |
| He et al. | Design and implementation of a unified MOOC recommendation system for social work major: Experiences and lessons | |
| Saroha et al. | Analysis and evaluation of learning management system using data mining techniques | |
| Sukhija et al. | Educational data mining towards knowledge engineering: a review state | |
| Xu | Digital Construction of Vocal Music Teaching Resource Base Using Data Mining Technology | |
| Wu | A music teaching resource management model based on fuzzy clustering algorithm | |
| Bin | Cognitive Web Service-Based Learning Analytics in Education Systems Using Big Data Analytics. | |
| CN110826590B (en) | Learner relationship strength measurement method and device integrating learning characteristics and learning network structural characteristics | |
| Taylor et al. | Rankings are the sorcerer’s new apprentice | |
| CN113657766A (en) | Tourist attraction joy index metering method based on tourist multi-metadata | |
| Rekha et al. | Ontology driven framework for assessing the syllabus fairness of a question paper | |
| CN112699933A (en) | Automatic identification method and system for processing capacity of user teaching material | |
| Kuosa et al. | Content analysis and visualizations: Tools for a social media-enhanced learning environment | |
| Zhou et al. | The practice and reform of classroom teaching based on the coordination of online environment and deep learning | |
| Hayashi et al. | Prediction of group learning results from an aggregation of individual understanding with kit-build concept map | |
| CN110909252A (en) | Community activity recommendation method and device based on big data | |
| Rajalakshmi | ANALYSIS AND PREDICTIONS ON BLENDED LEARNING READINESS AMONG INDIAN STUDENTS AT UNIVERSITIE SUSING DECISION TREE CLASSIFIER IN SCIKIT-LEARN ENVIRONMENT. | |
| Alvarez-Rodríguez et al. | Measuring the degree of innovation in education through the implementation of an indicator based on analyzing MOOC-related activities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190730 Address after: 310006 Room 410, 319 Shenjia Road, Xiacheng District, Hangzhou City, Zhejiang Province Applicant after: HANGZHOU BRAINS FOUNDATION TECHNOLOGY CO.,LTD. Address before: 310012 Room 502, Unit 1, 17 Jintian Garden Building, Xihu District, Hangzhou City, Zhejiang Province Applicant before: Yang Ying |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190820 |