CN107423343A - A kind of library book based on mixing collaborative filtering recommends method and system - Google Patents
A kind of library book based on mixing collaborative filtering recommends method and system Download PDFInfo
- Publication number
- CN107423343A CN107423343A CN201710334128.5A CN201710334128A CN107423343A CN 107423343 A CN107423343 A CN 107423343A CN 201710334128 A CN201710334128 A CN 201710334128A CN 107423343 A CN107423343 A CN 107423343A
- Authority
- CN
- China
- Prior art keywords
- books
- user
- neighbour
- matrix
- collaborative filtering
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24573—Query processing with adaptation to user needs using data annotations, e.g. user-defined metadata
-
- 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
- G06F16/21—Design, administration or maintenance of databases
- G06F16/211—Schema design and management
- G06F16/212—Schema design and management with details for data modelling support
-
- 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
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- 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
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Library & Information Science (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The present invention provides a kind of library book based on mixing collaborative filtering and recommends method, including following four steps：Step 1：Obtain book information data set；Step 2：Data prediction；Step 3：Establish user's books Rating Model；Step 4：Build user neighbour matrix and books neighbour's matrix；Step 5：Recommended Books；Same or similar for foundation with user interest and books by using mixing Collaborative Filtering Recommendation Algorithm, excavating user may be interested in which books.Book recommendation is carried out using this method, can more efficiently improve the utilization rate of books and reference materials, it is ensured that user obtains more effective resource.
Description
Technical field
The present invention relates to Information application and Internet technical field, more particularly to a kind of books based on mixing collaborative filtering
Shop book recommendation method and system.
Background technology
Library is the primary location that colleges and universities convey knowledge, and possesses substantial amounts of book information, newspapers and periodicals, user browse
Daily record etc., also inevitably result in the situation of library's information overload, and the overload of information is all brought to user and keeper
Certain puzzlement, user can only be by browsing Home Page or the lookup of data carried out by search engine, and in sea
Need to carry out substantial amounts of search work in the library of amount just to find the data of needs.But with information technology and interconnection
The development of network technology, the generation of commending system, user can targetedly obtain the data of correlation.Commending system and search are drawn
Cooperating with each other for holding up is effectively reduced the overload problem of data age.It is different from search engine main flow and uses PageRank algorithms,
It is directed to different application scenarios, commending system algorithm numerous and complicated, very different, ununified optimal algorithm.Compare
In the commending system of large-scale service website, its data set is often based upon historical record, and data are numerous and jumbled, input manpower and materials compared with
Greatly, it is not appropriate for library in the school.
The content of the invention
In view of this, recommend method the embodiment provides a kind of library book for facilitating taking care of books and be
System.
A kind of library book based on mixing collaborative filtering recommends method, comprises the following steps：Step 1：Obtain books
Message data set, the book information data set include user to the score data of books, reading daily record, books and the member of user
The metadata of data, the books and user include user's sex, age and book name, author, publishing house etc.；Step 2：Number
Data preprocess, the process of data preprocessing to the book information data set including carrying out data cleansing, default value is handled, different
Often processing, data conversion etc., carry out preference division；Step 3：Using the user to the score data of books according to
Preference quantifies to be assigned to corresponding weights structure user-books rating matrix, and is obtained by the user-books rating matrix
Go out user-books Rating Model；Step 4：Based on the user-books rating matrix structure user neighbour matrix and books neighbour
Matrix, the user neighbour matrix and books neighbour matrix determine user and user, figure according to the user-books rating matrix
Similarity numerical value between book and books is established；Step 5：Based on the user neighbour matrix and books neighbour's matrix, using mixing
Collaborative filtering recommending mode Recommended Books, draw final recommendation list.
Further, in the step 2, unessential user and books is deleted using the method for singular value decomposition, borrowed
This reduces the dimension of rating matrix.
Further, in the step 3, the preference is set to ten grades, and grade is that 1 corresponding weights are
0.1, grade is that 10 corresponding weights are 1, and the rest may be inferred.
Further, in the step 4, the similarity numerical value between the user and user, books and books is based on public affairs
Formula is calculated, the formula:
Try to achieve the similarity between user u and user v.ru,iTable
Show scorings of the user u to books i,The average that user u scores all books is represented, books i ∈ I, I represent books sum
Amount；
Try to achieve the similarity between books i and books j.ru,iTable
Show scorings of the user u to books i,The average value that all users score books i is represented, user u ∈ U, U represent total number of users
Amount.
Further, in the step 5, the mixing collaborative filtering recommending mode is multiple neighbour users and multiple near
The collaborative filtering mode that adjacent books combine, while train the multiple neighbour user and multiple neighbour's books both neighbour's moulds
Type.
Further, in the step 5, using the mode of cluster, neighbour user and the books of certain attribute is clustered, are obtained
To new neighbour.
Further, in the step 5, according to the evaluation and test predictor formula improved, quantitative calculating speculates user couple
The preference of certain this books, the evaluation and test predictor formula：
User v is to certain this figure
Book j fancy grade.Wherein
Parameter lambda is introduced to represent in the degree of dependence for mixing the collaborative filtering based on user and based on books, conuRepresent based on user's
The Confidence of collaborative filtering, coniRepresent the Confidence of the collaborative filtering based on books.
A kind of library book commending system based on mixing collaborative filtering, it is characterised in that：Including book information data
Collection acquiring unit, data pre-processing unit, user-books Rating Model establish unit, user neighbour matrix and books neighbour's square
Battle array construction unit, book recommendation unit, each unit are sequentially connected, and book information data set acquiring unit is used for book information data
Collection obtains；Data pre-processing unit is used for the pretreatment of the book information data set, including at data cleansing, default value
Reason, abnormality processing, data conversion etc.；User-books Rating Model is established data after unit is based on the pretreatment and is assigned to accordingly
User-books rating matrix constructed by weights establishes user-books Rating Model；User neighbour matrix and books neighbour's matrix
Construction unit is used to build based on user in the user-books rating matrix and user, the books number of degrees similar between books
It is worth user neighbour matrix and books neighbour's matrix；Book recommendation unit combines multiple neighbour users and multiple neighbour's books
Collaborative filtering mode be applied to the user neighbour matrix and books neighbour's matrix, draw final Recommended Books.
The beneficial effect brought of technical scheme that embodiments of the invention provide is：The present invention can be improved more efficiently
The utilization rate of books and reference materials, it is ensured that user obtains more effective resource.
Brief description of the drawings
Fig. 1 is that a kind of library book based on mixing collaborative filtering of the present invention recommends method flow diagram.
Fig. 2 is that a kind of library book based on mixing collaborative filtering of the present invention recommends the system of method to form figure.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to embodiment party of the present invention
Formula is further described.
Fig. 1 is refer to, the embodiment provides a kind of library book recommendation side based on mixing collaborative filtering
Method, this method comprise the following steps：
Step 1：Obtain book information data set.Data, including user are extracted from library's historical data to books
The metadata of scoring, the metadata for reading daily record, books and user, books and user include user's sex, age and books name
Title, author, publishing house etc..
Step 2：Data prediction.The Exploring Analysis that data are carried out according to the data of acquisition is handled, and carries out the pre- place of data
Reason, including data cleansing, default value processing, abnormality processing, data conversion.Using the method for singular value decomposition, solves scoring number
According to it is too sparse, find user neighbour when inaccuracy the problem of.The main purpose of data cleansing is the angle from analysis,
Filter out the data of needs.Because in data processing, the sex of user, age etc. do not influence the recommendation of books, reduce data
Redundancy during processing.Because partial data is default, therefore default value processing is carried out, it is former with reference to actual conditions and default value processing
Then, this partial data is directly filtered out.The wrong data concentrated in abnormality processing using data processing software to data is repaiied
Just.In data conversion, user profile is replaced with ID for protection privacy of user, while the travel log of user is handled, is entered
Row preference divides.
Step 3：Establish user-books Rating Model.The preference of user in step 2 is divided into ten grades, etc.
Level is 0.1 for 1 corresponding weights, and grade is that 10 corresponding weights are 1, and the rest may be inferred.So as to obtain user as follows-books scoring
R in matrix R, rating matrix Ru,iEvaluations of the user u to books i is represented, row vector represents that a certain user is commented all books
Valency, column vector represent that all users to a certain evaluation of books, thus construct user-books model.
Step 4：Build user neighbour matrix and books neighbour's matrix.Specific steps：
Step 401, the scoring list of each user is regarded as a dimension vector, then can be according to the similar of vector
It is similar between degree quantification user.When searching has k user of similar preference to books, select be based on Pearson came phase here
Similarity between the calculating user of relation number.According to formulaTry to achieve user u and
Similarity between user v.Similarity between obtained user u and each user is generated into similarity matrix with this from high to low
User-Similarity[U][N]；
Wherein, ru,iScorings of the user u to books i is represented,Represent the average that user u scores all books, books
I ∈ I, I represent books total quantity.
Step 402, the scoring list of each user is regarded as a dimension vector, then can be according to the similar of vector
It is similar between degree quantification user.When searching there are the approximate books of k sheets of preference for user, select be based on Pierre here
The improved similarity between calculating books of inferior formula of correlation coefficient.According to formula
The similarity between books i and books j is tried to achieve, obtained books i and similarity between each books are generated into phase with this from high to low
Like degree matrix Item-Similarity [I] [M]；
Wherein, ru,iScorings of the user u to books i is represented,Represent the average value that all users score books i, user
U ∈ U, U represent total number of users amount.
Step 403, the Similarity value between the user obtained according to step 401 carries out the division of user's nearest-neighbor, makes
With selected Fix-size, the sequencing of similarity that is obtained according to Fix-size chooses the Optimal units of more neighborhoods, or according to
Similarity value is more than defined threshold value and is classified as a neighborhood by defined threshold.Will be similar between obtained each user with reference to field
Degree generates similarity matrix User-Similarity [U] [N] with this from high to low, and wherein U is total number of users amount, and N is that user is near
Adjacent number；The division that the Similarity value between books carries out books nearest-neighbor, obtained figure are similarly obtained according to step 302
Similarity between book generates similarity matrix Item-Similarity [I] [M] with this from high to low, and wherein I is books sum
Amount, M are neighbour's number in books；
Step 5：Recommended Books.The way of recommendation is mixed using the collaborative filtering based on user and books, draws final push away
Recommend list.Specific steps：
Step 501, according to User-Similarity [U] [N] the user similarity matrix and Item- generated in step 4
Similarity [I] [M] books similarity matrix, can simultaneously directly in a matrix retrieval can obtain faster user v and
This books of N number of user and I of books j arest neighbors.It is different from in post-processing stages mixing based on user and based on books
Both collaborative filterings the obtained prediction result of method, be separate between its method.Multiple neighbour users and multiple
The collaborative filtering method that neighbour's books combine trains two kinds of Neighborhood Models simultaneously so that they can be mutual in learning parameter
Solution, and reduce space complexity.
Step 502, this books of user u and books i N number of user and I of arest neighbors have been obtained according to step 501, but by
In corresponding neighbour user partly lacking evaluation neighbour's books, therefore in this step using the mode of cluster, according to certain
Hierarchical cluster attribute neighbour user and books, obtain new user and books neighbour.The user neighbour such as obtained concentrates, and has user's figure
Book review valency lacks, and user neighbour is collected into cluster dividing using cluster, cluster is represented with center neighbour's data, obtained cluster is exactly new near
Successive term.
Step 503, recommendation calculating is carried out according to new neighbour, draws best books.Improve evaluation and test predictor formulaQuantitative calculating, which speculates, to be used
Fancy grades of the family v to certain this books j.Wherein
Parameter lambda is introduced to represent in the degree of dependence for mixing the collaborative filtering based on user and based on books, conuRepresent base
In the Confidence of the collaborative filtering of user, coniThe Confidence of the collaborative filtering based on books is represented, U and I represent user and Tu
The arest neighbors collection of book.
Fig. 2 is refer to, the library book commending system based on mixing collaborative filtering can be formed using the above method, including
It is near that book information data set acquiring unit 10, data pre-processing unit 11, user-books Rating Model establish unit 12, user
Adjacent matrix and books neighbour matrix construction unit 13, book recommendation unit 14, book information data set acquiring unit 10 are used to scheme
Book message data set obtains；Data pre-processing unit 11 is used for the pretreatment of the book information data set, including data is clear
Wash, default value processing, abnormality processing, data conversion etc.；User-books Rating Model establish unit 12 be used for establish be based on described in
The user of pretreated data-books rating matrix；User neighbour matrix and books neighbour matrix construction unit 13 are used for structure
Build and user's neighbour's matrix is drawn based on the similarity in the user-books rating matrix between user and user, books and books
With books neighbour's matrix；The collaborative filtering mode that book recommendation unit 14 combines multiple neighbour users and multiple neighbour's books should
For the user neighbour matrix and books neighbour's matrix, final Recommended Books are drawn.
Claims (8)
1. a kind of library book based on mixing collaborative filtering recommends method, it is characterised in that comprises the following steps：Step 1：
Obtain book information data set, the book information data set include user to the score datas of books, read daily record, books and
The metadata of the metadata of user, the books and user include user's sex, age and book name, author, publishing house etc.；
Step 2：Data prediction, the process of data preprocessing include carrying out data cleansing, default to the book information data set
Value processing, abnormality processing, data conversion etc., carry out preference division；Step 3：Utilize scoring number of the user to books
User-books rating matrix is built according to corresponding weights are assigned to according to preference quantization, and by the user-books
Rating matrix draws user-books Rating Model；Step 4：Based on the user-books rating matrix structure user's neighbour's matrix
With books neighbour's matrix, the user neighbour matrix and books neighbour matrix determine according to the user-books rating matrix
Similarity numerical value between user and user, books and books is established；Step 5：Based on the user neighbour matrix and books neighbour
Matrix, using mixing collaborative filtering recommending mode Recommended Books, draw final recommendation list.
2. a kind of library book based on mixing collaborative filtering as claimed in claim 1 recommends method, it is characterised in that：
In the step 2, unessential user and books are deleted using the method for singular value decomposition, reduce the dimension of rating matrix.
3. a kind of library book based on mixing collaborative filtering as claimed in claim 1 recommends method, it is characterised in that：
In the step 3, the preference is set to ten grades, and grade is that 1 corresponding weights are 0.1, and grade is 10 corresponding power
It is worth for 1, the rest may be inferred.
4. a kind of library book based on mixing collaborative filtering as claimed in claim 1 recommends method, it is characterised in that：
In the step 4, the similarity numerical value between the user and user, books and books is drawn based on formula, the formula:
Try to achieve the similarity between user u and user v, ru,iRepresent to use
Scorings of the family u to books i,The average that user u scores all books is represented, books i ∈ I, I represent books total quantity；
Try to achieve the similarity between books i and books j, ru,iRepresent to use
Scorings of the family u to books i,The average value that all users score books i is represented, user u ∈ U, U represent total number of users amount.
5. a kind of library book based on mixing collaborative filtering as claimed in claim 1 recommends method, it is characterised in that：
In the step 5, the mixing collaborative filtering recommending mode is the collaboration of multiple neighbour users and the combination of multiple neighbour's books
Filter mode, while train the multiple neighbour user and multiple both Neighborhood Models of neighbour's books.
6. a kind of library book based on mixing collaborative filtering as claimed in claim 1 recommends method, it is characterised in that：
In the step 5, using the mode of cluster, neighbour user and the books of certain attribute are clustered, obtain new neighbour.
7. a kind of library book based on mixing collaborative filtering as claimed in claim 1 recommends method, it is characterised in that：
In the step 5, according to the evaluation and test predictor formula improved, quantitative calculating speculates preference of the user to certain this books,
The evaluation and test predictor formula：
Try to achieve v couples of user
Certain this books j fancy grade, wherein
Parameter lambda is introduced to represent in the degree of dependence for mixing the collaborative filtering based on user and based on books, conuRepresent based on user's
The Confidence of collaborative filtering, coniRepresent the Confidence of the collaborative filtering based on books.
A kind of 8. library book commending system based on mixing collaborative filtering, it is characterised in that：Including book information data set
Acquiring unit, data pre-processing unit, user-books Rating Model establish unit, user neighbour matrix and books neighbour's matrix
Construction unit, book recommendation unit, each unit are sequentially connected, and book information data set acquiring unit is used for book information data set
Obtain；Data pre-processing unit is used for the pretreatment of the book information data set, including data cleansing, default value processing,
Abnormality processing, data conversion etc.；User-books Rating Model establishes data after unit is based on the pretreatment and is assigned to corresponding weight value
Constructed user-books rating matrix establishes user-books Rating Model；User neighbour matrix and books neighbour matrix structure
Unit is used to build to be obtained based on the similarity numerical value in the user-books rating matrix between user and user, books and books
Go out user neighbour matrix and books neighbour's matrix；The association that book recommendation unit combines multiple neighbour users and multiple neighbour's books
It is applied to the user neighbour matrix and books neighbour's matrix with filter type, draws final Recommended Books.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710334128.5A CN107423343A (en) | 2017-05-12 | 2017-05-12 | A kind of library book based on mixing collaborative filtering recommends method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710334128.5A CN107423343A (en) | 2017-05-12 | 2017-05-12 | A kind of library book based on mixing collaborative filtering recommends method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107423343A true CN107423343A (en) | 2017-12-01 |
Family
ID=60425461
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710334128.5A Pending CN107423343A (en) | 2017-05-12 | 2017-05-12 | A kind of library book based on mixing collaborative filtering recommends method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107423343A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304556A (en) * | 2018-02-06 | 2018-07-20 | 中国传媒大学 | The personalized recommendation method being combined with collaborative filtering based on content |
| CN108491412A (en) * | 2018-01-31 | 2018-09-04 | 广东易联创富集团有限公司 | Children's book recommends method, apparatus, platform, computer readable storage medium |
| CN108510373A (en) * | 2018-04-12 | 2018-09-07 | 京东方科技集团股份有限公司 | Paintings recommend method, paintings recommendation apparatus, equipment and storage medium |
| CN109460518A (en) * | 2018-12-07 | 2019-03-12 | 杭州东信北邮信息技术有限公司 | A kind of book recommendation method based on user website access record |
| CN109907351A (en) * | 2019-03-21 | 2019-06-21 | 杭州电子科技大学 | A kind of cigarette composition maintenance method and system based on mixing collaborative filtering |
| CN110334281A (en) * | 2019-07-11 | 2019-10-15 | 广东工业大学 | A kind of book recommendation method, device, equipment and the medium of combination user behavior |
| CN110490786A (en) * | 2019-07-10 | 2019-11-22 | 广东工业大学 | A kind of books update method based on distributed intelligence books station |
| CN110659423A (en) * | 2019-09-19 | 2020-01-07 | 辽宁工程技术大学 | School side learning material recommendation method based on collaborative filtering |
| CN111259236A (en) * | 2020-01-09 | 2020-06-09 | 贵州大学 | Recommendation method for donation crowd funding field |
| CN116720927A (en) * | 2023-08-08 | 2023-09-08 | 北京人天书店集团股份有限公司 | Book recommendation method, system and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104391849A (en) * | 2014-06-30 | 2015-03-04 | 浙江大学苏州工业技术研究院 | Collaborative filtering recommendation method for integrating time contextual information |
| CN105183727A (en) * | 2014-05-29 | 2015-12-23 | 上海研深信息科技有限公司 | Method and system for recommending book |
| US20160275081A1 (en) * | 2013-03-20 | 2016-09-22 | Nokia Technologies Oy | Method and apparatus for personalized resource recommendations |
| CN106202184A (en) * | 2016-06-27 | 2016-12-07 | 华中科技大学 | A kind of books personalized recommendation method towards libraries of the universities and system |
-
2017
- 2017-05-12 CN CN201710334128.5A patent/CN107423343A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160275081A1 (en) * | 2013-03-20 | 2016-09-22 | Nokia Technologies Oy | Method and apparatus for personalized resource recommendations |
| CN105183727A (en) * | 2014-05-29 | 2015-12-23 | 上海研深信息科技有限公司 | Method and system for recommending book |
| CN104391849A (en) * | 2014-06-30 | 2015-03-04 | 浙江大学苏州工业技术研究院 | Collaborative filtering recommendation method for integrating time contextual information |
| CN106202184A (en) * | 2016-06-27 | 2016-12-07 | 华中科技大学 | A kind of books personalized recommendation method towards libraries of the universities and system |
Non-Patent Citations (1)
| Title |
|---|
| 宋宇辰，孟海东 编著: "5.2.2 数据预处理", 《数据挖掘与管理实践》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108491412A (en) * | 2018-01-31 | 2018-09-04 | 广东易联创富集团有限公司 | Children's book recommends method, apparatus, platform, computer readable storage medium |
| CN108304556A (en) * | 2018-02-06 | 2018-07-20 | 中国传媒大学 | The personalized recommendation method being combined with collaborative filtering based on content |
| CN108510373A (en) * | 2018-04-12 | 2018-09-07 | 京东方科技集团股份有限公司 | Paintings recommend method, paintings recommendation apparatus, equipment and storage medium |
| CN109460518B (en) * | 2018-12-07 | 2020-07-24 | 杭州东信北邮信息技术有限公司 | Book recommendation method based on user website access records |
| CN109460518A (en) * | 2018-12-07 | 2019-03-12 | 杭州东信北邮信息技术有限公司 | A kind of book recommendation method based on user website access record |
| CN109907351A (en) * | 2019-03-21 | 2019-06-21 | 杭州电子科技大学 | A kind of cigarette composition maintenance method and system based on mixing collaborative filtering |
| CN110490786A (en) * | 2019-07-10 | 2019-11-22 | 广东工业大学 | A kind of books update method based on distributed intelligence books station |
| CN110334281A (en) * | 2019-07-11 | 2019-10-15 | 广东工业大学 | A kind of book recommendation method, device, equipment and the medium of combination user behavior |
| CN110334281B (en) * | 2019-07-11 | 2022-02-15 | 广东工业大学 | Book recommendation method, device, equipment and medium combining user behaviors |
| CN110659423A (en) * | 2019-09-19 | 2020-01-07 | 辽宁工程技术大学 | School side learning material recommendation method based on collaborative filtering |
| CN111259236A (en) * | 2020-01-09 | 2020-06-09 | 贵州大学 | Recommendation method for donation crowd funding field |
| CN116720927A (en) * | 2023-08-08 | 2023-09-08 | 北京人天书店集团股份有限公司 | Book recommendation method, system and storage medium |
| CN116720927B (en) * | 2023-08-08 | 2023-11-03 | 北京人天书店集团股份有限公司 | Book recommendation method, system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107423343A (en) | A kind of library book based on mixing collaborative filtering recommends method and system | |
| CN103870973B (en) | Information push, searching method and the device of keyword extraction based on electronic information | |
| CN101320375B (en) | Digital book search method based on user click action | |
| CN103455487B (en) | The extracting method and device of a kind of search term | |
| CN105528437B (en) | A kind of question answering system construction method extracted based on structured text knowledge | |
| CN111708740A (en) | Mass search query log calculation analysis system based on cloud platform | |
| CN110968782A (en) | Student-oriented user portrait construction and application method | |
| CN104951518B (en) | One kind recommends method based on the newer context of dynamic increment | |
| CN110674407A (en) | Hybrid recommendation method based on graph convolution neural network | |
| CN103646092A (en) | SE (search engine) ordering method based on user participation | |
| CN105653840A (en) | Similar case recommendation system based on word and phrase distributed representation, and corresponding method | |
| CN101894351A (en) | Multi-agent based tour multimedia information personalized service system | |
| CN107038184B (en) | A kind of news recommended method based on layering latent variable model | |
| CN104298785B (en) | Searching method for public searching resources | |
| CN109101493A (en) | A kind of intelligence house-purchase assistant based on dialogue robot | |
| CN102855282B (en) | A kind of document recommendation method and device | |
| CN113761383B (en) | Search and recommendation fusion system based on unified user behavior modeling | |
| CN112966091B (en) | Knowledge map recommendation system fusing entity information and heat | |
| CN108717411B (en) | Questionnaire design auxiliary system based on big data | |
| CN107330627A (en) | A kind of big data processing method, server and system for innovating intention | |
| US20170235836A1 (en) | Information identification and extraction | |
| CN102004772A (en) | Method and equipment for sequencing search results according to terms | |
| CN110970112B (en) | Knowledge graph construction method and system for nutrition and health | |
| CN111475724A (en) | Random walk social network event recommendation method based on user similarity | |
| Zubiaga et al. | Content-based clustering for tag cloud visualization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Xu Hongwen Inventor after: Li Ling Inventor after: Chen Wen Inventor after: Xue Yinxi Inventor after: Yin Weiming Inventor after: Xie Jing Inventor after: Chen Zhenhua Inventor after: Zeng Hao Inventor before: Li Ling Inventor before: Xue Yinxi Inventor before: Chen Zhenhua Inventor before: Zeng Hao |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171201 |