CN107016566A - User model construction method based on body - Google Patents
User model construction method based on body Download PDFInfo
- Publication number
- CN107016566A CN107016566A CN201710117737.5A CN201710117737A CN107016566A CN 107016566 A CN107016566 A CN 107016566A CN 201710117737 A CN201710117737 A CN 201710117737A CN 107016566 A CN107016566 A CN 107016566A
- Authority
- CN
- China
- Prior art keywords
- user
- interest
- model
- information
- degree
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 19
- 230000008569 process Effects 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 abstract description 2
- 230000009467 reduction Effects 0.000 abstract description 2
- 230000004048 modification Effects 0.000 description 9
- 238000012986 modification Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 6
- 239000013598 vector Substances 0.000 description 5
- 230000008901 benefit Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000007935 neutral effect Effects 0.000 description 2
- 230000037396 body weight Effects 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000001568 sexual effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Classifications
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/14—Travel agencies
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of user model construction method based on body, comprise the following steps：(1) relevant information of user characteristics, preference and demand can be reflected by obtaining；(2) tour field body is built；(3) user is classified using similarity algorithm；(4) interest model is built to each user, user model is represented using body；(5) mode for transmitting adjustment using interest-degree is updated to user model.The present invention collects the essential information of user based on Ontological concept hierarchical tree by web usage logs, using these information and field ontology library domain body is set up using Ontology Editing Tool, user is classified through row by improved similarity algorithm, and user model is set up to every class user, finally user interest model is updated with reference to the method for interest-degree and transmission adjustment, so that the structure difficulty reduction of model, has higher confidence level and the degree of accuracy to description user interest.
Description
Technical field
The present invention relates to the modeling method of tour field, more specifically to a kind of user model structure based on body
Construction method.
Background technology
Tourist industry, which is one, has concentrated the comprehensive industries of the links such as traffic, visit, lodging, food and drink, shopping, entertainment.With
Developing rapidly for network technology, people's retrievable travel information on network shows a kind of explosive growth situation,
These network information resources bring enrich one's knowledge with great convenience while, can also be caused information to individual fatigue
And information pressure.In order to be quickly and accurately positioned the travel information of needs, personalized ground pushed information is given people, it is emerging to user
The research of interesting model has become focus and difficult point instantly.At present, some common user model structure sides have been occurred in that
Formula：Based on keyword, based on vector space model, neutral net, Evaluations matrix etc., but these user models are all
In the presence of certain defect.Mode less stable based on vector space, as a result usually there is many deviations；Based on neutral net
Method is not then readily understood, and the scope of application is smaller；Adaptability based on Evaluations matrix is poor, it is difficult to accomplish to update interest；Therefore,
These models can not accurately describe the personal interest of user.
Because body has certain advantage in user interest description, gradually by many scholar's research.Research direction bag
Include：(1) cognitive structure of user how is described using bulk form, is that user builds the user model based on domain body,
Improve the quality of personalized retrieval；(2) user model is built based on user knowledge body and Concept Vectors, realizes user interest
Personalized semantic description；(3) the weighting body of integrated user interest information and semantic information is built, and is carried based on the weighting body
Property semantic search framework is one by one gone out；(4) updated by user personality body, correction body and body and realize user's mould
The structure of type, in being studied more than, domain knowledge is a domain classification system, and user interest is emerging to each concept theme
Interest still employs the description form of weighted keywords；(5) made with the class of ontology extraction Web community users sessions, attribute and example
It is characterized item, feature based occurrence frequency, semantic locations and interest model more new algorithm calculating characteristic item in domain body
Weight；(6) it is the concept hierarchy that node and side are constituted by ontology representation, each node is related to the document sets for representing its content
Connection, calculates document sets weight by tf-idf methods and generates knot vector, calculate institute's directed quantity in advance by index entry；(7) ternary
Group representation：Description, interest-degree and the last update time of Ontological concept；(8) quadruple notation method is proposed：User property
Concept set and user are to set of relations between the visit capacity of each concept, user in collection, domain body, domain body；(9) hexa-atomic group is proposed
Representation：In user personalized information (interest model mark), user interest Ontological concept collection, user interest degree collection, concept set
Paired Concept Semantic Similarity, concept creation time, project the last time accessed time contained by concept.
But, retouching for crucial term vector is still employed for resource during the study and renewal of user interest model
State form.Although applying domain knowledge, basic technology is still based on the describing mode and traditional machine of keyword
The application of device learning algorithm, and most of the user model set up a simply concept hierarchy, user model is quiet
State, it is impossible to which with the change of user interest, dynamic upgrades in time, and the stability of user model is relatively low.User model can not be with
User interest change (containing content change, old interest attenuation, the generation of new interest) and improvement in time through row adaptability, from
And the process of interests change can not be reflected well.
The content of the invention
In view of the shortcomings of the prior art, it is an object of the invention to provide a kind of user model structure side based on body
Method, preferably can be positioned to resource, improve travel information retrieval performance.
To achieve these goals, the technical scheme is that：A kind of user model construction method based on body, bag
Include following steps：(1) relevant information of user characteristics, preference and demand can be reflected by obtaining；(2) tour field body is built；
(3) user is classified using similarity algorithm；(4) model is built to each user, user's mould is represented using body
Type；(5) mode for transmitting adjustment using interest-degree is updated to user model.
As a modification of the present invention, in step (1), the acquisition modes of described information include explicit collection and implicit
Collection, explicit collection directly collects the essential information of user by the log-on message of user；Implicitly gather by analyzing user
Web usage logs are recorded, and essential information inf, the user for obtaining user are general to the preference v and access body of leaf node
The visit capacity V of leaf node in hierarchical tree is read, wherein, shown in the preference such as formula (a) of user, shown in visit capacity such as formula (b)：
V=vi (1<=i<=n) represent user's degree interested in i-th of leaf node, fI, rRepresent that user accesses leaf
The number of times of child node Li people's resource, n represents the number of leaf node in Ontological concept hierarchical tree, according to the access of leaf node
Amount calculates the visit capacity of non-leaf nodes, the preference information of user is supplemented complete.
As a modification of the present invention, in step (2), building the process of tour field body includes：Set up file,
Set up class and its class hierarchy, set up attribute and the permissible value of attribute, addition example.
As a modification of the present invention, in step (3), the formula of the similarity algorithm is：
Sim (i, j)=w × rij/es(i,j)+(1-w)×S(i,j) 0<w<1
Wherein, w is corresponding weight, rijThe quantity accessed jointly for similar users with not accessed jointly by the two users
Resource quantity ratio, S (i, j) represent resource between visit capacity similarity.
As a modification of the present invention, in step (4), the interest model of user is a triple, includes tourism neck
Domain body, user basic information and user interest body, the essential information of user are stored in Top nodes, and Top is user
The top layer concept of model, the mark and the essential information of user of the browsed travel information of its object set record user, it belongs to
Property include node attribute and user to the interest-degree of certain node, interest-degree is used as numerical attribute deposit property set.
As a modification of the present invention, in step (5), based on the interest evolution Model for forget-encouraging process again,
Fill that user increases newly in time originally without interest, rejecting some uninterested reached after certain threshold value of user in time
The interest-degree of interest, the in time original interest of user that adjustment level of interest is changed.
Compared with prior art, the present invention collects the basic of user based on Ontological concept hierarchical tree by web usage logs
Information, sets up domain body using Ontology Editing Tool using these information and field ontology library, passes through improved similarity operator
Method is classified to user through row, and sets up user model to every class user, finally combine the method for interest-degree and transmission adjustment to
Family interest model is updated so that the structure difficulty reduction of model, has higher confidence level and accurate to description user interest
Degree.
Brief description of the drawings
With reference to the accompanying drawings and detailed description, the structure and its advantageous effects to the present invention are carried out specifically
It is bright.
Fig. 1 is tourism Ontology Modeling flow chart of the invention.
Fig. 2 is tourism ontology model figure of the invention.
Embodiment
In order that goal of the invention, technical scheme and its advantageous effects of the present invention become apparent from, below in conjunction with accompanying drawing
And embodiment, the present invention will be described in further detail.It should be appreciated that the specific reality described in this specification
Apply mode to explain the present invention, be not intended to limit the present invention.
Good concept hierarchy and the support to reasoning from logic that body has, the node of body hierarchical tree is one
Individual keyword or metadata, from tree root to leaf, the problem of refinement of concept is only existed between node, and in the absence of node scale
Less and less the problem of.User model based on body can be illustrated as having the form of the concept map of level, and be stored in
In general relational database, information retrieval is completed using the matching technique of figure.User model based on body is typically used
A kind of description language (such as Loom) represents that user model is stored in knowledge base, passes through the reasoning from logic energy of description language
Power completes information retrieval.The structure of user model based on body, is exactly represented using the concept in body or body
User model, is either represented using body or Ontological concept, will a clear and definite problem, that is, user model is to rely on
Exist in body.If representing a user model using a body, i.e., each user model corresponds to a body,
Then user model just contains the relation between concept and concept, when using user model, it is possible to according to user model
Itself calculate, without inquiring about body in itself, still, it is necessary to repair during the relation modification standardized present in the body
Change the relation stored in each user model, generate the Redundancy of data.If being represented using the concept in body
User model, i.e., represent the interest of user using one group of interest concept, it ensure that interest conceptual relation and body in system
The uniformity of concept, still, is fed back, for systematicness when each user interest model updates, it is necessary to inquire about body
It can have a certain impact.The present invention utilizes the father in the conceptual level number of times of body on the basis of the domain body having been built up
Subrelation and brotherhood, form a user model for being similar to number form shape.The present invention is by body with a triple
It is indicated, is tour field body, user basic information and user interest body respectively.
Fig. 1 and Fig. 2 are referred to, the user model construction method of the invention based on body comprises the following steps：
(1) user profile is collected and processing.Because the difference of user context causes its interest worlds also to have their own characteristics each, for example
University student may select the higher self-service trip of sexual valence, and top managers can then be more likely to quality trip, thus firstly the need of to
Classified at family.The collection of user data is the process for obtaining the relevant information that can reflect user characteristics, preference and demand, this
The data of body can be divided into explicit collection and implicit acquisition mode.The explicit main log-on message by user of collection, is directly received
Collect the essential information of user.Implicitly collection mainly analysis user web usage logs are recorded, the essential information inf of acquisition user,
User is to the preference v of leaf node and accesses the visit capacity V of leaf node in Ontological concept hierarchical tree, wherein, user
Preference such as formula (a) shown in, shown in visit capacity such as formula (b).
V=vi (1<=i<=n) represent user's degree interested in i-th of leaf node.fI, rRepresent that user accesses leaf
The number of times of child node Li people's resource, n represents the number of leaf node in Ontological concept hierarchical tree, according to the access of leaf node
Amount calculates the visit capacity of non-leaf nodes so that the preference information supplement of user is complete.
(2) structure of tour field body.Tour field body is built using the OWL language in Protege instruments.
In protege, building the process of domain body includes setting up file, sets up class and its class hierarchy, set up permitting for attribute and attribute
Perhaps it is worth, adds example this 4 basic steps.After tourism body frame has been built, it is necessary to instance data is added under framework,
The example exploitation of class in body also makes constantly to add and perfect by iterative cycles.To give full play to body in knowledge table
Advantage in terms of showing with reasoning from logic, it would be desirable to the semantic relation between conscientious analysis classes and class, and between attribute
Some related relations.The domain knowledge base built using ontology not only can clearly describe concept in field and its
Relation, can also realize the shared of domain knowledge and reuse, and be conducive to the management and maintenance of domain knowledge base.
(3) tourism user classification.User profile includes the personal essential information of user, mainly name, sex, the age,
Occupation, level of consumption etc..Calculated herein by the user's visit capacity and the essential information of user in Ontological concept hierarchical tree
User's similitude, and then many and complicated users are classified through row.Because its different tour interest field of user context also can
Difference, the tour interest of young and old people also has difference.Personalization system passes through the user's information that is collected into
Classified.Generally user's node interested is the sub-fraction in Ontological concept hierarchical tree, using improved
Similarity algorithm is classified to user：Sim (i, j)=w × rij/es(i,j)+(1-w)×S(i,j)0<w<1, wherein, w is phase
The weight answered, rijThe quantity ratio of the quantity accessed jointly for similar users and the resource not accessed jointly by the two users, s
(i, j) represents the visit capacity similarity between resource.Similar user can be obtained according to improved similarity formula, so that
Realize that user classifies.
(4) ontology representation of user's tour interest model.User is realized after classification, interest mould is built to each user
Type, i.e., represent user model using body, and the building process of User-ontology model is to be directed to different users, using constructed
Domain body in node the process of user interest described.Add and use on the part of nodes of the domain body built
The personal information at family is that can obtain user model, and the interest model of user is considered as a triple, i.e. tour field body,
User basic information and user interest body.When using ontology representation user model, the essential information of user is stored in Top
In node, Top is the top layer concept of user model, the mark of the browsed all travel informations of its object set record user and
Some essential informations of user, the attribute of its attribute including node and user are used as numerical value to the interest-degree of certain node, interest-degree
Attribute is stored in property set.The interest preference information of the detailed description user of user model energy, it is ensured that the service of personalization system
Quality.
(5) renewal of user model.With the growth of age of user, the increase of social experience, the influence of work etc. is more
The relation of weight.Over time, its interest is constantly changing, and for the interest of accurate real reflection user, just needs
To carry out timely modification to user interest to update, the mode for transmitting adjustment using interest-degree is updated user model.This
Invention using based on forgets-encouraging the interest evolution Model of process again, user increased newly originally without interest need benefit in time
On, user loses interest in some original interest, and reaching needs rejecting in time after certain threshold value, and user is believed existing interest
The level of interest of breath is changed, it is necessary to be adjusted to corresponding interest-degree.
User is indicated by the present invention to the preference of tour interest with body, to more fully understand that the behavior of user is inclined
It is good, by setting up the configuration file of a user model, the user profile corresponding to the resource that user is liked is deposited, user
The user profile corresponding to resource not liked, and the user profile corresponding to general resource.Body enters to user preference resource
The abstract of row user profile, the weights proportion of one resource information of each body correspondence.Abstract is carried out with body can be more preferably
User behavior information is described on ground.Resource is ranked up by the weight shared by resource, passed through again after accessing every time
Specific algorithm carries out weights identification and user profile identification again to resource, preferably can be positioned to resource, in trip
It is improved in terms of trip information retrieval performance (precision ratio and recall ratio).
The announcement and teaching of book according to the above description, those skilled in the art in the invention can also be to above-mentioned embodiment party
Formula carries out appropriate change and modification.Therefore, the invention is not limited in embodiment disclosed and described above, to this
Some modifications and changes of invention should also be as falling into the scope of the claims of the present invention.Although in addition, this specification
In used some specific terms, but these terms are merely for convenience of description, do not constitute any limitation to the present invention.
Claims (6)
1. a kind of user model construction method based on body, it is characterised in that comprise the following steps：
(1) relevant information of user characteristics, preference and demand can be reflected by obtaining；
(2) tour field body is built；
(3) user is classified using similarity algorithm；
(4) model is built to each user, user model is represented using body；
(5) mode for transmitting adjustment using interest-degree is updated to user model.
2. the user model construction method according to claim 1 based on body, it is characterised in that in step (1), institute
Stating the acquisition modes of information includes explicit collection and implicit collection, and explicit collection is directly collected and used by the log-on message of user
The essential information at family；Implicit collection is recorded by analyzing user web usage logs, obtains essential information inf, the user couple of user
The preference v of the leaf node and visit capacity V for accessing leaf node in Ontological concept hierarchical tree, wherein, the preference of user
Shown in degree such as formula (a), shown in visit capacity such as formula (b)：
V=vi (1<=i<=n) user's degree interested in i-th of leaf node is represented, fi, r represents that user accesses leaf
The number of times of node Li people's resource, n represents the number of leaf node in Ontological concept hierarchical tree, according to the visit capacity of leaf node
The visit capacity of non-leaf nodes is calculated, the preference information of user is supplemented complete.
3. the user model construction method according to claim 1 based on body, it is characterised in that in step (2), structure
Building the process of tour field body includes：File is set up, class is set up and its class hierarchy, sets up the permissible value of attribute and attribute, adds
Plus example.
4. the user model construction method according to claim 1 based on body, it is characterised in that described in step (3)
The formula of similarity algorithm is：
Sim (i, j)=w × rij/eS (i, j)+(1-w)×S(i,j)0<w<1
Wherein, w is corresponding weight, rijThe quantity accessed jointly for similar users and the money not accessed jointly by the two users
The quantity ratio in source, S (i, j) represents the visit capacity similarity between resource.
5. the user model construction method according to claim 1 based on body, it is characterised in that in step (4), is used
Family model is a triple, includes tour field body, user basic information and user interest body, the essential information of user
It is stored in Top nodes, Top is the top layer concept of user model, the mark of the browsed travel information of its object set record user
Know and user essential information, the attribute of its attribute including node and user are used as number to the interest-degree of certain node, interest-degree
Value attribute is stored in property set.
6. the user model construction method according to claim 1 based on body, it is characterised in that in step (5), base
In the interest evolution Model for forget-encouraging process again, fill that user increases newly in time originally without interest, reject use in time
Uninterested some interest reached after certain threshold value in family, the user that adjustment level of interest is changed in time is original emerging
The interest-degree of interest.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710117737.5A CN107016566A (en) | 2017-03-01 | 2017-03-01 | User model construction method based on body |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710117737.5A CN107016566A (en) | 2017-03-01 | 2017-03-01 | User model construction method based on body |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107016566A true CN107016566A (en) | 2017-08-04 |
Family
ID=59440820
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710117737.5A Pending CN107016566A (en) | 2017-03-01 | 2017-03-01 | User model construction method based on body |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107016566A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107392812A (en) * | 2017-09-01 | 2017-11-24 | 安徽教育网络出版有限公司 | A kind of education of middle and primary schools resource application service system based on semanteme |
| CN107507041A (en) * | 2017-09-08 | 2017-12-22 | 北京京东尚科信息技术有限公司 | The construction method and construction device of user model |
| CN109409916A (en) * | 2017-08-18 | 2019-03-01 | 徐子明 | A kind of Tourism Marketing system based on big data platform |
| CN110738493A (en) * | 2018-07-19 | 2020-01-31 | 上海交通大学 | Body maintenance system based on block chain |
| CN113377926A (en) * | 2021-06-28 | 2021-09-10 | 中国标准化研究院 | Construction method of registration meta-model of quality information ontology evolution |
-
2017
- 2017-03-01 CN CN201710117737.5A patent/CN107016566A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409916A (en) * | 2017-08-18 | 2019-03-01 | 徐子明 | A kind of Tourism Marketing system based on big data platform |
| CN109409916B (en) * | 2017-08-18 | 2021-10-22 | 重庆赫皇科技咨询有限公司 | Tourism marketing system based on big data platform |
| CN107392812A (en) * | 2017-09-01 | 2017-11-24 | 安徽教育网络出版有限公司 | A kind of education of middle and primary schools resource application service system based on semanteme |
| CN107507041A (en) * | 2017-09-08 | 2017-12-22 | 北京京东尚科信息技术有限公司 | The construction method and construction device of user model |
| CN110738493A (en) * | 2018-07-19 | 2020-01-31 | 上海交通大学 | Body maintenance system based on block chain |
| CN110738493B (en) * | 2018-07-19 | 2023-04-18 | 上海交通大学 | Body maintenance system based on block chain |
| CN113377926A (en) * | 2021-06-28 | 2021-09-10 | 中国标准化研究院 | Construction method of registration meta-model of quality information ontology evolution |
| CN113377926B (en) * | 2021-06-28 | 2022-11-25 | 中国标准化研究院 | Construction method of registration meta-model of quality information ontology evolution |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Qi et al. | Finding all you need: web APIs recommendation in web of things through keywords search | |
| CN104933164B (en) | In internet mass data name entity between relationship extracting method and its system | |
| CN107016566A (en) | User model construction method based on body | |
| CN109189944A (en) | Personalized recommending scenery spot method and system based on user's positive and negative feedback portrait coding | |
| CN110555112A (en) | interest point recommendation method based on user positive and negative preference learning | |
| Lin et al. | Heterogeneous knowledge-based attentive neural networks for short-term music recommendations | |
| CN105045875B (en) | Personalized search and device | |
| CN112966091B (en) | Knowledge map recommendation system fusing entity information and heat | |
| CN108710663A (en) | A kind of data matching method and system based on ontology model | |
| CN104834686A (en) | Video recommendation method based on hybrid semantic matrix | |
| El Mohadab et al. | Predicting rank for scientific research papers using supervised learning | |
| CN108664658A (en) | A kind of collaborative filtering video recommendation method considering user preference dynamic change | |
| CN105320719A (en) | Crowdfunding website project recommendation method based on project tag and graphical relationship | |
| CN116975615A (en) | Task prediction method and device based on video multi-mode information | |
| Zhong et al. | Design of a personalized recommendation system for learning resources based on collaborative filtering | |
| CN109949174A (en) | A kind of isomery social network user entity anchor chain connects recognition methods | |
| CN112948710A (en) | Graph neural network-based punishment education recommendation method, system and storage medium | |
| CN107562966A (en) | The optimization system and method based on intelligence learning for web page interlinkage retrieval ordering | |
| CN115374781A (en) | Text data information mining method, device and equipment | |
| Miao | Mobile Information System of English Teaching Ability Based on Big Data Fuzzy K‐Means Clustering | |
| CN118069927A (en) | News recommendation method and system based on knowledge perception and user multi-interest feature representation | |
| CN113934835A (en) | Retrieval type reply dialogue method and system combining keywords and semantic understanding representation | |
| CN102193928B (en) | Method for matching lightweight ontologies based on multilayer text categorizer | |
| CN117171428B (en) | Method for improving accuracy of search and recommendation results | |
| CN114329167A (en) | Hyper-parameter learning, intelligent recommendation, keyword and multimedia recommendation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170804 |
|
| RJ01 | Rejection of invention patent application after publication |