CN107194012A - A kind of good courseware commending system of recommendation effect - Google Patents
A kind of good courseware commending system of recommendation effect Download PDFInfo
- Publication number
- CN107194012A CN107194012A CN201710489329.2A CN201710489329A CN107194012A CN 107194012 A CN107194012 A CN 107194012A CN 201710489329 A CN201710489329 A CN 201710489329A CN 107194012 A CN107194012 A CN 107194012A
- Authority
- CN
- China
- Prior art keywords
- node
- courseware
- user
- power
- holding power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06F16/26—Visual data mining; Browsing structured data
-
- 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
Abstract
The invention provides a kind of good courseware commending system of recommendation effect, the system end for passing through network connection including client terminal and with the client terminal, the system end includes member and excavates subsystem and courseware recommendation subsystem, the usage mining subsystem is used to excavate influential user from appling courseware person, and the courseware recommends subsystem to be used for influential user and recommends courseware to client terminal.Beneficial effects of the present invention are：Courseware recommendation is carried out using influential user, the courseware's quality of recommendation is higher.
Description
Technical field
The present invention relates to courseware recommended technology field, and in particular to a kind of good courseware commending system of recommendation effect.
Background technology
In current Network Multimedia Teaching field, the content of courses, this side are provided to client more with businessman's way of recommendation
Formula can not serve the current Web-based instruction market increasingly changed well.Carry out recommending to be easier to allow client to produce by user
It is raw to trust.
Annexation of the user network between individual and individual is constituted.Individual be also referred to as node, can be tissue, individual,
The entity or virtual individual of the difference implication such as network ID；And interaction among individuals can be blood relationship, it is cooperation, alliance, hostile
Etc. various relations.Choose influential user and carry out the key point that courseware is recommended to recommend as solution user.
The content of the invention
In view of the above-mentioned problems, a kind of the present invention is intended to provide good courseware commending system of recommendation effect.
The purpose of the present invention is realized using following technical scheme：
There is provided a kind of good courseware commending system of recommendation effect, including client terminal and pass through with the client terminal
The system end of network connection, the system end includes member and excavates subsystem and courseware recommendation subsystem, usage mining
System is used to excavate influential user from appling courseware person, and the courseware recommends subsystem to be used for influential user
Recommend courseware to client terminal.
Beneficial effects of the present invention are：Courseware recommendation is carried out using influential user, the courseware's quality of recommendation is higher.
Brief description of the drawings
Using accompanying drawing, the invention will be further described, but the embodiment in accompanying drawing does not constitute any limit to the present invention
System, for one of ordinary skill in the art, on the premise of not paying creative work, can also be obtained according to the following drawings
Other accompanying drawings.
Fig. 1 is the structural representation of the present invention；
Fig. 2 is present system end structure schematic diagram.
Reference：
System end 1, client terminal 2, member excavate subsystem 11, courseware and recommend subsystem 12.
Embodiment
The invention will be further described with the following Examples.
Referring to Fig. 1, Fig. 2, a kind of good courseware commending system of recommendation effect of the present embodiment, including client terminal 2 and
With system end 1 of the client terminal 2 by network connection, the system end 1 includes member's excavation subsystem 11 and courseware is pushed away
Subsystem 12 is recommended, the usage mining subsystem 11 is used to excavate influential user, the courseware from appling courseware person
Recommend subsystem 12 to be used for influential user and recommend courseware to client terminal 2.
The present embodiment carries out courseware recommendation using influential user, and the courseware's quality of recommendation is higher.
It is preferred that, the courseware recommends subsystem 12 to include a courseware database, and the influential user is from courseware
Recommend courseware to client terminal 2 in database.
This preferred embodiment sets up courseware database, is recommended from courseware database, more can efficiently carry out
Recommend.
It is preferred that, the client terminal 2 is computer or mobile phone.
It is convenient that this preferred embodiment is realized, no matter user stay at home or road on can easily read courseware.
It is preferred that, the usage mining subsystem 11 includes user network modeling module, influence power analysis module and user
Module is excavated, the user network modeling module is used to set up user network model, and the influence power analysis module is used for basis
User network model is analyzed node influence power, and the usage mining module is used to carry out user according to node influence power
Excavate；Specifically the user network model is set up in the following ways：Regard each user as a node, use triple G=
(V, E, EH) represents user network, wherein, EH represents the set of each node initial value, EHG(v) represent node v in user network
Initial value on network G, v ∈ V represent node, and V represents node set, and e ∈ E represent the relation between node, and E is represented between node
Annexation set.
Effective excavation of user in this preferred embodiment usage mining subsystem realizes appling courseware person, specifically, will
User network is modeled as trigram models, has not only completely taken out user network, and model simple, intuitive, can be convenient
Useful information is obtained, initial value can be accomplished in several ways as an abstract function, come for example with PageRank
Obtain initial value.
It is preferred that, the influence power analysis module include the first dependency degree calculating sub module, the second holding power calculating sub module and
3rd influence power analyzes submodule, and the first dependency degree calculating sub module is used for the dependency degree of calculate node, and described second supports
Power calculating sub module is used for the holding power of the dependency degree calculate node according to node, and the 3rd influence power analysis submodule is used for root
The influence power of node is analyzed according to the holding power of node；The dependency degree of specific calculate node in the following ways：For any
Node u, v ∈ V, node u are expressed as to v dependency degree：
In above-mentioned formula, EHG(u) initial values of the node u in user network G is represented, EM (u → v) represents dependences of the node u to v
Degree, G-v represent user network G disconnected nodes v be connected with other nodes after user network；Dependency degree is smaller, relationships between nodes
More become estranged, dependency degree is bigger, relationships between nodes are closer.
This preferred embodiment influence power analysis module sets the first dependency degree calculating sub module to ask for the dependency degree of node, leads to
Cross and set up dependency degree function, obtain the close and distant relation of user network interior joint, calculated for subsequent node holding power and established good
Good basis, in dependency degree calculating process, calculates dependency degree, it is ensured that in user network by the way of disconnected node is connected
Interstitial content is constant, considers EHG-vAnd EH (u)G(u) calculate dependency degree, obtain the higher dependency degree of confidence level, for
Family holding power is calculated and laid the foundation.
It is preferred that, the second holding power calculating sub module includes a holding power computing unit and secondary holding power is calculated
Unit, a holding power computing unit is used for the first holding power of calculate node, and the secondary holding power computing unit is used
In the second holding power of calculate node；First holding power of calculate node is in the following ways：(1) user network is converted into two
Tuple G '=(V, EM), EM represent the dependency degree set between node, for node u, and v ∈ V, EM (u → v) represent node u to v's
Dependency degree；(2) in G ', the set that v relies on maximum node is combined into for v ∈ V, v arest neighbors collection, is represented with N (v), v's
Reverse Nearest collection is combined into the set for relying on v maximum node, is represented with RN (v)；(3) in G=(V, EM), for u ∈
V, then u the first holding power be represented by：Above-mentioned formula
In, YW1(u) node u the first holding power is represented, | N (v) | represent v arest neighbors interstitial contents；Second holding power of calculate node
In the following ways：(1) user network is converted into two tuple G '=(V, EM), EM represents the dependency degree set between node, right
In node u, v ∈ V, EM (u → v) represent dependency degrees of the node u to v；(2) in G ', v is combined into for v ∈ V, v k neighbours collection
To node dependency degree in the set of preceding k node, represented with kN (v), v Reverse Nearest collection is combined into v dependency degrees preceding k's
The set of node, is represented with kRN (v)；(3) in G=(V, EM), for u ∈ V, then u the second holding power is represented by：In above-mentioned formula, YW2(u) node u the second holding power is represented, |
KN (v) | represent v k neighboring node numbers.
This preferred embodiment influence power analysis module sets the second holding power calculating sub module to ask for the holding power of node, leads to
The first holding power and the second holding power of operator node respectively are crossed, more comprehensively node is obtained and supports force information, for follow-up
Node influence power is calculated and had laid a good foundation, wherein, the first holding power is calculated based on nearest neighbors, the second holding power base
Calculated in k neighboring nodes, obtained holding power of the node under different application environment.
It is preferred that, specifically the influence power in the following ways to node is analyzed：Set up the influence force function of node：In above-mentioned formula, LG (u) represents node u influence power
Functional value, influence power functional value is bigger, shows that the influence power of node is bigger；Specifically carry out digging user in the following ways
Pick：The influence power functional value of user is calculated, n maximum node of influence power functional value is chosen and is used as usage mining result.
This preferred embodiment influence power analysis module sets the 3rd influence power to analyze submodule calculate node influence power, passes through
Influence force function is set up, the first holding power and the second holding power of node, the node influence power more section of acquisition has been considered
Learn accurate, be that the follow-up important individual found in user network or colony provide guarantee, so as to ensure that courseware is recommended
Quality；Usage mining module in usage mining subsystem is excavated according to influence power functional value to user, obtained use
Family more meets demand, so as to preferably carry out courseware recommendation.
Courseware recommendation is carried out using the good courseware commending system of recommendation effect of the present invention, when courseware recommended amount be respectively 1,
2nd, 3,4,5 when, recommend time and CSAT to count courseware, compared with the existing technology, the beneficial effect of generation is such as
Shown in following table：
| Courseware recommended amount | The recommendation time shortens | CSAT is improved |
| 1 | 23% | 21% |
| 2 | 25% | 20% |
| 3 | 30% | 25% |
| 4 | 27% | 19% |
| 5 | 20% | 23% |
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than to present invention guarantor
The limitation of scope is protected, although being explained with reference to preferred embodiment to the present invention, one of ordinary skill in the art should
Work as understanding, technical scheme can be modified or equivalent substitution, without departing from the reality of technical solution of the present invention
Matter and scope.
Claims (7)
1. a kind of good courseware commending system of recommendation effect, it is characterised in that including client terminal and with the client terminal
By the system end of network connection, the system end includes member and excavates subsystem and courseware recommendation subsystem, and the user digs
Pick subsystem is used to excavate influential user from appling courseware person, and it is influential that the courseware recommends subsystem to be used for
User recommends courseware to client terminal.
2. the good courseware commending system of recommendation effect according to claim 1, it is characterised in that the courseware recommends subsystem
System includes a courseware database, and the influential user recommends courseware from courseware database to client terminal.
3. the good courseware commending system of recommendation effect according to claim 2, it is characterised in that the client terminal is meter
Calculation machine or mobile phone.
4. the good courseware commending system of recommendation effect according to claim 3, it is characterised in that the usage mining subsystem
System 11 includes user network modeling module, influence power analysis module and usage mining module, and the user network modeling module is used
In setting up user network model, the influence power analysis module is used to divide node influence power according to user network model
Analysis, the usage mining module is used to excavate user according to node influence power；Specifically set up in the following ways described
User network model：Regard each user as a node, user network is represented with triple G=(V, E, EH), wherein, EH tables
Show the set of each node initial value, EHG(v) initial values of the node v on user network G is represented, v ∈ V represent node, V
Node set is represented, e ∈ E represent the relation between node, and E represents the annexation set between node.
5. the good courseware commending system of recommendation effect according to claim 4, it is characterised in that the influence power analyzes mould
Block includes the first dependency degree calculating sub module, the second holding power calculating sub module and the 3rd influence power analysis submodule, described the
One dependency degree calculating sub module is used for the dependency degree of calculate node, and the second holding power calculating sub module is used for according to node
The holding power of dependency degree calculate node, the 3rd influence power analysis submodule is for shadow of the holding power according to node to node
Power is rung to be analyzed；The dependency degree of specific calculate node in the following ways：For arbitrary node u, v ∈ V, node u to v according to
Lai Du is expressed as：In above-mentioned formula, EHG(u) section is represented
Initial values of the point u in user network G, EM (u → v) represents dependency degrees of the node u to v, and G-v represents that user network G disconnects
Node v be connected with other nodes after user network；Dependency degree is smaller, and relationships between nodes are more become estranged, and dependency degree is bigger, between node
Relation is closer.
6. the good courseware commending system of recommendation effect according to claim 5, it is characterised in that the second holding power meter
Operator module includes a holding power computing unit and secondary holding power computing unit, and a holding power computing unit is used for
First holding power of calculate node, the secondary holding power computing unit is used for the second holding power of calculate node；Calculate node
The first holding power in the following ways：(1) user network is converted into two tuple G '=(V, EM), EM represent between node according to
Lai Du gathers, for node u, and v ∈ V, EM (u → v) represent dependency degrees of the node u to v；(2) in G ', for v ∈ V, v most
Neighbour's collection is combined into the set that v relies on maximum node, is represented with N (v), and v Reverse Nearest collection, which is combined into, relies on v maximum section
The set of point, is represented with RN (v)；(3) in G=(V, EM), for u ∈ V, then u the first holding power is represented by：In above-mentioned formula, YW1(u) first of node u is represented
Holding force, | N (v) | represent v arest neighbors interstitial contents；Second holding power of calculate node is in the following ways：(1) by user network
Two tuple G '=(V, EM) is converted into, EM represents the dependency degree set between node, section is represented for node u, v ∈ V, EM (u → v)
Dependency degrees of the point u to v；(2) in G ', for v ∈ V, v k neighbours collection be combined into v to node dependency degree preceding k node collection
Close, represented with kN (v), v Reverse Nearest collection is combined into v dependency degrees in the set of preceding k node, is represented with kRN (v)；(3)
In G=(V, EM), for u ∈ V, then u the second holding power is represented by：
In above-mentioned formula, YW2(u) represent that the second of node u supports
Power, | kN (v) | represent v k neighboring node numbers.
7. the good courseware commending system of recommendation effect according to claim 6, it is characterised in that specifically in the following ways
Influence power to node is analyzed：Set up the influence force function of node： In above-mentioned formula, LG (u) represents node u influence power functional value, and influence power functional value is bigger, shows section
The influence power of point is bigger；Specifically carry out excavating user in the following ways：The influence power functional value of user is calculated, is chosen
N maximum node of influence power functional value is used as usage mining result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710489329.2A CN107194012A (en) | 2017-06-24 | 2017-06-24 | A kind of good courseware commending system of recommendation effect |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710489329.2A CN107194012A (en) | 2017-06-24 | 2017-06-24 | A kind of good courseware commending system of recommendation effect |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107194012A true CN107194012A (en) | 2017-09-22 |
Family
ID=59878325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710489329.2A Pending CN107194012A (en) | 2017-06-24 | 2017-06-24 | A kind of good courseware commending system of recommendation effect |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107194012A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040260688A1 (en) * | 2003-06-05 | 2004-12-23 | Gross John N. | Method for implementing search engine |
| CN102156706A (en) * | 2011-01-28 | 2011-08-17 | 清华大学 | Mentor recommendation system and method |
| CN102456064A (en) * | 2011-04-25 | 2012-05-16 | 中国人民解放军国防科学技术大学 | Method for realizing community discovery in social networking |
| CN104408210A (en) * | 2014-12-31 | 2015-03-11 | 合一网络技术(北京)有限公司 | Video recommendation method based on opinion leaders |
| CN105959365A (en) * | 2016-04-26 | 2016-09-21 | 中国联合网络通信集团有限公司 | Application recommendation method and application recommendation device |
| CN106874509A (en) * | 2017-03-01 | 2017-06-20 | 广州大学 | Resource recommendation method and device based on middle granularity user grouping |
-
2017
- 2017-06-24 CN CN201710489329.2A patent/CN107194012A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040260688A1 (en) * | 2003-06-05 | 2004-12-23 | Gross John N. | Method for implementing search engine |
| CN102156706A (en) * | 2011-01-28 | 2011-08-17 | 清华大学 | Mentor recommendation system and method |
| CN102456064A (en) * | 2011-04-25 | 2012-05-16 | 中国人民解放军国防科学技术大学 | Method for realizing community discovery in social networking |
| CN104408210A (en) * | 2014-12-31 | 2015-03-11 | 合一网络技术(北京)有限公司 | Video recommendation method based on opinion leaders |
| CN105959365A (en) * | 2016-04-26 | 2016-09-21 | 中国联合网络通信集团有限公司 | Application recommendation method and application recommendation device |
| CN106874509A (en) * | 2017-03-01 | 2017-06-20 | 广州大学 | Resource recommendation method and device based on middle granularity user grouping |
Non-Patent Citations (1)
| Title |
|---|
| 韩毅: "社会网络分析与挖掘的若干关键问题研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102750336B (en) | Resource individuation recommendation method based on user relevance | |
| Opricovic et al. | Extended VIKOR method in comparison with outranking methods | |
| CN102646103B (en) | The clustering method of term and device | |
| CN105101093B (en) | A kind of network topology method for visualizing on geographical location information | |
| CN103617279A (en) | Method for achieving microblog information spreading influence assessment model on basis of Pagerank method | |
| CN102262681A (en) | Method for identifying key blog sets in blog information spreading | |
| CN101887441A (en) | Method and system for establishing social network and method and system for mining network community | |
| CN104361036A (en) | Association rule mining method for alarm event | |
| Coclite et al. | Numerical schemes for computing discontinuous solutions of the Degasperis–Procesi equation | |
| CN104680250A (en) | Position predicting system | |
| CN104598648B (en) | A kind of microblog users interactive mode gender identification method and device | |
| CN104462443B (en) | Data processing method and device | |
| CN105678590A (en) | topN recommendation method for social network based on cloud model | |
| CN101976279A (en) | Method for evaluating engineering structure finite element model | |
| CN108153917A (en) | A kind of real estate investigation banking process for visualizing with defining people room relationship | |
| Zhang et al. | Joint intelligence ranking by federated multiplicative update | |
| Narayanamoorthy et al. | Adaptation of induced fuzzy cognitive maps to the problems faced by the power loom workers | |
| CN107194012A (en) | A kind of good courseware commending system of recommendation effect | |
| CN110247805A (en) | A kind of method and device for propagating key node based on the identification that K shell decomposes | |
| Mashat et al. | Discovery of association rules from university admission system data | |
| CN113723072A (en) | RPA (resilient packet Access) and AI (Artificial Intelligence) combined model fusion result acquisition method and device and electronic equipment | |
| Lin et al. | PRO‐GRADE: GIS toolkits for ground water recharge and discharge estimation | |
| CN106817271B (en) | Method and device for forming flow map | |
| Chen et al. | No TVD fields for 1-d isentropic gas flow | |
| Yu et al. | Distributed H∞ filtering with consensus in sensor networks: a two-dimensional system-based approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170922 |