CN103093376A - Clustering collaborative filtering recommendation system based on singular value decomposition algorithm - Google Patents
Clustering collaborative filtering recommendation system based on singular value decomposition algorithm Download PDFInfo
- Publication number
- CN103093376A CN103093376A CN2013100163818A CN201310016381A CN103093376A CN 103093376 A CN103093376 A CN 103093376A CN 2013100163818 A CN2013100163818 A CN 2013100163818A CN 201310016381 A CN201310016381 A CN 201310016381A CN 103093376 A CN103093376 A CN 103093376A
- Authority
- CN
- China
- Prior art keywords
- user
- cluster
- commodity
- matrix
- recommendation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a clustering collaborative filtering recommendation technology based on a singular value decomposition algorithm. The clustering collaborative filtering recommendation technology based on the singular value decomposition algorithm comprises firstly classifying users by using user attributive character values provided by the clustering collaborative filtering recommendation technology based on the singular value decomposition algorithm, and reducing dimension of a user-commodity grade matrix; improving a singular value decomposition (SVD) algorithm which is frequently used in image processing and natural language processing, and using the improved SVD algorithm in a recommendation system; decomposing a grade matrix in a cluster where users are located, and aggregating the decomposed grade matrix so as to fill predicted scores of non-grade items in the grade matrix, calculating similarity of the users in the same cluster by using the filled grade matrix, calculating final predicted scores of a commodity by applying collaborative filtering technologies based on the users and widely applied in the recommendation system, and carrying out final recommendation. The clustering collaborative filtering recommendation technology based on the singular value decomposition algorithm has the advantages of being capable of improving recommendation efficiency of the recommendation system, solving the problems such as data sparsity of the recommendation system, and meanwhile being capable of improving accuracy rate of recommendation of the recommendation system.
Description
Technical field
The invention belongs to the Technologies of Recommendation System in E-Commerce field, be specifically related to integrated multiple technologies, as data digging technology, machine learning techniques, natural language processing technique etc., realize a kind of recommend method that cluster is combined with svd (SVD) technology.
Background technology
In recent years, along with the fast development of Internet technology, ecommerce has become a kind of new fashion, forms the trend that rapidly increases in recent years.Ecommerce, it is that the IT technology is combined a kind of new business transaction process that produces with commercial act, is the Main Patterns of 21 century market economy business operating, by e-commerce platform, people can enjoy the quick and convenient of the home-confined free choice of goods.Expansion along with the e-commerce platform transaction size, people can't browse all commodity at short notice fast by browser, and also lack some product introductions that in the reality transaction, the shop-assistant carries out client, so people have faced electronic commerce times distinctive " information overload " problem.
Be directed to " information overload " problem, commending system arose at the historic moment in the nineties in 20th century, as the news of Google recommend, Email filters etc.Present nearly all e-commerce system, all with the inevitable ingredient of recommended technology as online spending, as Amazon, Netflix, bean cotyledon, Taobao etc.The Main Function of commending system has: (1) induces new client, namely to a potential new lead referral product, the viewer is become the buyer; (2) encourage the frequent customer, namely recommend more products on the basis of the thing that the client has bought, improve the cross-selling ability of network; (3) promote client to the loyalty of website.Accuracy rate, extensibility, real-time are to estimate a commending system quality whether key factor.Yet along with people more and more hanker after shopping online, present commending system is faced with the problem of " information overload ", because the data volume in system is too huge, causes some present recommended technologies can not effectively make real-time recommendation; Meanwhile, a problem that is always perplexing commending system is exactly local data's sparse property problem, although the data volume in system is very big, but for each sole user, its browse with the commodity of buying shared systems in the ratio all too of total commodity number little, this just can't accomplish accurate and effective on the problem that causes calculating user's similarity, has affected so greatly the result of recommending.Therefore how solving above-mentioned two problems becomes the subject matter that commending system needs to be resolved hurrily.
This patent adopts the cluster collaborative filtering recommending technology based on singular value decomposition algorithm, by the user is first classified, reduce the dimension of user-commodity rating matrix, then utilize svd (SVD) algorithm, polymerization after in the cluster of user place, rating matrix being decomposed, thereby the prediction score of the not scoring item in filled matrix is made final recommendation by collaborative filtering at last.This technology can improve system recommendation efficient, solves the problem such as commending system Deta sparseness and can improve the recommendation accuracy rate of system.
Summary of the invention
The present invention proposes to improve by the technology of utilizing clustering algorithm to combine with svd (SVD) algorithm recommendation efficient and the accuracy rate of commending system.Cause traditional commending system recommend the lower main cause of efficient be due to the number of users that exists in commending system and commodity amount too much, and traditional commending system must be by calculating the similarity between every two users or commodity, find out with designated user or the highest k the arest neighbors of commodity similarity and make recommendation, because calculated amount is too huge, and be not that each time calculating is all necessary, therefore cause recommending Efficiency Decreasing; Although causing traditional commending system to recommend the lower main cause of accuracy rate is because the data volume in a system is very huge, but for each sole user, its browse with the commodity of buying shared systems in the ratio all too of total commodity number little, the sparse property of this local data just can't be accomplished accurate and effective on the problem that causes calculating user's similarity, has affected so greatly the accuracy rate of recommending.The present invention solves by clustering algorithm and recommends efficiency, and utilizes svd algorithm to solve local data's sparse property problem, thereby improves recommendation results.
Technical solution of the present invention is divided into following several basic execution in step:
Step 1: utilize the user characteristics value, comprise sex, age, occupation, existing subscriber in system is assigned in n cluster go, and calculate the cluster centre value of each cluster;
Step 2: for the user of each new registration, calculate this user attribute characteristic values, should value and each cluster centre value compare, find out a most close cluster, should new user join, and change the cluster centre value of this cluster;
Step 3: for each login user, find this user place cluster, user in this cluster-commodity rating matrix is taken out, and according to must the rule, utilize svd algorithm that this rating matrix is decomposed, polymerization again after matrix after decomposing is processed, thus the prediction that obtains each commodity of not marking in original matrix is marked;
Step 4: utilize the new rating matrix that obtains, calculate the similarity of each user in this login user and this matrix, find out the highest k of a similarity user as his arest neighbors;
Step 5: utilize the arest neighbors similarity based on the login user that obtains in user's collaborative filtering method and step 4, each commodity of not marking in original rating matrix are carried out score in predicting, obtain at last best recommendation results, and this result is recommended the user.
The present invention has following technical characterictics:
(1) the described clustering algorithm of step 1 is to utilize k-means clustering algorithm principle famous in data mining, a kind of new clustering algorithm that is applicable to commending system has been proposed in the present invention, by utilizing the sex in user's registration information, the attributes such as age and occupation, utilize the many experiments data to find out the weights of each optimum attributive character, thereby draw a formula that calculates user's synthesized attribute eigenwert, carrying out cluster according to this user's synthesized attribute eigenwert;
(2) the described process of step 2 is exactly the process of constantly adding new data in the system database in fact, by the new user's synthesized attribute of the calculating eigenwert after registration, new user is carried out cluster, in the time of can be afterwards each user's login, find fast this user place cluster, improve the efficient of system;
(3) the described svd algorithm of step 3 is mainly used to the problem of the sparse property of resolution system local data.Svd algorithm is generally used for the fields such as image processing and natural language processing, the present invention with this method improvement after, use in recommended technology and go, by calculating for the first time the prediction scoring of the commodity of not marking in user-commodity rating matrix, solve the problem of the sparse property of local data of system.
(4) the described process of step 4, not simple to mark to recommend with the prediction of the resulting commodity of not marking in step 3, but calculate user's similarity in same cluster by resulting scoring in step 3, divide rating matrix after depolymerizing by svd algorithm, make the calculating of user's similarity more accurate.
(5) the described process of step 5, it is recommendation part last in commending system of the present invention, find targeted customer's k arest neighbors by step 4, utilize traditional collaborative filtering recommending method based on the user, calculate the prediction scoring of the commodity of not marking in original subscriber-commodity matrix, and last result is recommended the targeted customer.
Embodiment
For making purpose of the present invention, technical scheme and advantage clearer, below will give an actual example is described in detail the present invention.
1. utilize the clustering algorithm of user attribute characteristic values classification
In the present invention, we utilize user's synthesized attribute eigenwert to carry out clustering to the user.Usually, each user has the personal characteristics of oneself, as wage, native place, sex, occupation, age etc.According to the well-known Ai Rui of consulting firm, Chinese netizen's consumption status is added up, consumer's consumption can be divided according to user's different attribute.And the information that common user can collect in the time of registration comprises sex, age and occupation, and these 3 kinds of attributes exactly can well reflect a user's feature.Therefore we can utilize above attribute that the user is classified, and by concrete computing formula are: muc (u)=aS (u)+β A (u)+γ O (u), calculate user's attributive character value.A+ β+γ=1 wherein, they are respectively the related coefficient of each attribute.The synthesized attribute eigenwert of muc (u) expression user u,
The sex character value of S (u) expression user u,
The age characteristics value of A (u) expression user u,
The job characteristics value of O (u) expression user u,
Draw user's n cluster by above formula, and for each new user, utilize formula muc (u)=aS (u)+β A (u)+γ O (u), this user place cluster is found, in order to next find this user's arest neighbors in this cluster in a+ β+γ=1 wherein, test by mass data and show, work as a=0.2, β=0.5, the Clustering Effect that γ=0.3 o'clock the method obtains is best.
If a system has 3 cluster A, B, C at present.The cluster centre attributive character value of A is 0.32, and in cluster, number of users is that the cluster centre attributive character value of 8, B is 0.65, and in cluster, number of users is that the cluster centre attributive character value of 9, C is 0.89, and in cluster, number of users is 10.A new user a registers in this system, and his sex is the man, and the age is 28 years old, and occupation is high school teachers, belongs to cultural class occupation subtree.So according to the appeal formula, we can obtain the attributive character value muc (a) of user a=0.2*1+0.5* (28-15)/40+0.3*0.5=0.51, should value do with 3 cluster centre values respectively and take absolute value after poor, find minimum one, find that this value and B cluster centre are the most approaching, therefore a is joined in cluster B, and the eigenwert of change B cluster centre is (9*0.65+0.51)/10=0.64.
2. svd (SVD) algorithm
Svd (Singular Value Decomposition, SVD) be a kind of matrix decomposition technology, it has disclosed the inner structure of matrix deeply, and svd has a wide range of applications at aspects such as compression of images, least square method, natural language processings.It can (the matrix R that supposes m 〉=n) be decomposed into three matrix U, S, V:R=U * S * V with a m * n
T, wherein U is the orthogonal matrix (UU of a m * m
T=I), V is the orthogonal matrix (VV of a n * n
T=I), S is the diagonal matrix of a m * n, element on its off-diagonal is 0 entirely, and the element on diagonal line satisfy σ 1 〉=σ 2 〉=... 〉=σ n 〉=0, all σ n are greater than 0 and arrange according to descending order, be called singular value (Singular Value), singular value can represent that is given the degree of closeness of set matrix with the matrix lower than its order.
3. utilize svd algorithm to decompose user-commodity rating matrix method
This method utilizes svd algorithm to solve in user-commodity matrix the not sparse property problem of scoring item, utilizes the rating matrix of polymerization after filling to calculate similarity between the user, thereby improves the recommendation accuracy of system.
Algorithm:
Input: initial score matrix R (R is the two-dimensional matrix of m * n, and m represents to have m user, and n represents to have n commodity, and the project of scoring is not filled to 0)
Output: each user u is added the matrix PR of original initial score matrix to the prediction scoring of scoring item i not
(1) obtain the average mark ri of the commodity scoring in each row;
(2) all replace to rui-ri with what in original matrix, each had a non-zero score value rui, and obtain matrix R ';
(3) utilize svd algorithm, R ' is resolved into U, S, three matrixes of V;
(4) S being simplified, the value less than 1 in all diagonal matrix S all is made as 0, because S diagonal matrix singular value is descending arrangement, is therefore 0 row and column deletion with all numerical value in s-matrix, generates new S
kDiagonal matrix (be equivalent to keep front K capable and front K row);
(5) according to the S that simplifies
k, corresponding to U, the V matrix reduction is U
K, V
KMatrix, U
KBe the two-dimensional matrix of m * k, V
KTwo-dimensional matrix for k * n;
(6) calculate S
kSquare root S
k 1/2, obtain respectively two matrix A=U
k* S
k 1/2B=S
k 1/2* V
k
(7) fill the PR matrix, for the project of not marking in each initial matrix R
(
Be the mean value of user u scoring, A
mThe m of representing matrix A is capable, B
nThe n row of representing matrix B);
4. utilize traditional collaborative filtering recommending method based on the user to calculate similarity and prediction score value
Classical collaborative filtering has following three kinds: common cosine similarity; The cosine similarity of Pearson coefficient and correction.The present invention adopts the Pearson's coefficient method to find the solution similarity between the user.Formula is as follows:
I wherein
uvThe project that expression user u and v marked jointly, PR
ui, PR
viBe illustrated respectively in user u and the v scoring to commodity i in the PR matrix;
With
Represent respectively user u and the v average to all commodity in the PR matrix.Produce because the PR matrix decomposes polymerization again through svd algorithm, so the user rolls up the score information of commodity, effectively solved the similarity that Deta sparseness causes and calculated inaccurate problem.According to the similarity that following formula is obtained, find out k and the highest user of user u similarity, utilize the prediction evaluate formula to predict scoring to the not scoring item in initial matrix R,
Here, N (u, k) expression and the most similar k of a user u interest user's set, U (i) comments undue user's set, R to article i
viBe user v to the scoring of article i,
With
Be respectively user v and user u all comment the scoring mean value of undue article.Calculate at last the commodity projection scoring that all are not marked, the result of optimum is recommended the user.
Claims (5)
1. based on the cluster Collaborative Filtering Recommendation System of singular value decomposition algorithm, it is characterized in that, before calculating user's similarity, at first each user in system is divided in different clusters according to specific clustering algorithm and goes, can effectively reduce like this dimension of user-commodity rating matrix, improve the recommendation efficient of system; Secondly will be usually used in image process with natural language processing technique in svd (SVD) algorithm be applied in commending system after improving and go, with polymerization again after user-commodity matrix decomposition, thereby draw the commodity scoring after prediction for the first time, calculate the similarity between the user, can effectively solve matrix Sparse Problems common in commending system like this.
2. method according to claim 1, is characterized in that, it is two steps that implementation is divided into, cluster and utilize svd algorithm to fill user-commodity matrix.Cluster is responsible for original a large number of users in system is divided in a plurality of little clusters and is gone; Utilize svd algorithm fill user-commodity matrix be responsible for in the user in the cluster of this user place-commodity rating matrix the item of scoring fill by polymerization after decomposing, utilize above two steps can reduce the system-computed dimension, the sparse property of resolution system problem, thus efficient and the accuracy rate of recommending greatly improved.
3. method according to claim 1, it is characterized in that, cluster is responsible for utilizing user's attributive character value that original a large number of users-merchandise news in system is divided in a plurality of little clusters and is gone, the user characteristics value mainly comprises: sex, age and occupation, by the test of many times data analysis, take out the shared weight of each eigenwert, thereby draw the cluster formula of an optimal computed synthesized attribute eigenwert.
4. method according to claim 1, is characterized in that, utilize svd algorithm fill user-commodity matrix be responsible for in the user in the cluster of this user place-commodity rating matrix the item of scoring fill by polymerization after decomposing.At first this step finds out user place cluster, utilize svd algorithm to decompose the user in cluster-commodity rating matrix, rear in polymerization by processing, thereby the commodity that obtain predicting for the first time scoring, utilize these prediction scorings can calculate the similarity of user between same cluster, reduce the sparse property of system data problem.Calculating between designated user place cluster user after similarity, utilizing to get similarity, original user-commodity rating matrix is being carried out score in predicting for the second time, thereby draw last recommendation results.
5. method according to claim 1, it is characterized in that, after a user registers login, system is by his sex, the attributive character value that age and occupation calculate the user finds this user place cluster, utilize svd algorithm to draw k the user the most similar to this user interest hobby, and by calculating the prediction scoring of commodity, this user is made accurately recommending at last.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310016381.8A CN103093376B (en) | 2013-01-16 | 2013-01-16 | Clustering collaborative filtering recommendation system based on singular value decomposition algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310016381.8A CN103093376B (en) | 2013-01-16 | 2013-01-16 | Clustering collaborative filtering recommendation system based on singular value decomposition algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103093376A true CN103093376A (en) | 2013-05-08 |
| CN103093376B CN103093376B (en) | 2020-02-14 |
Family
ID=48205914
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310016381.8A Active CN103093376B (en) | 2013-01-16 | 2013-01-16 | Clustering collaborative filtering recommendation system based on singular value decomposition algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103093376B (en) |
Cited By (46)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103279552A (en) * | 2013-06-06 | 2013-09-04 | 浙江大学 | Collaborative filtering recommendation method based on user interest groups |
| CN103324690A (en) * | 2013-06-03 | 2013-09-25 | 焦点科技股份有限公司 | Mixed recommendation method based on factorization condition limitation Boltzmann machine |
| CN104111969A (en) * | 2014-06-04 | 2014-10-22 | 百度移信网络技术（北京）有限公司 | Method and system for measuring similarity |
| CN104636371A (en) * | 2013-11-11 | 2015-05-20 | 华为技术有限公司 | Information recommendation method and device |
| CN104731887A (en) * | 2015-03-13 | 2015-06-24 | 东南大学 | User similarity measuring method in collaborative filtering |
| CN104751842A (en) * | 2013-12-31 | 2015-07-01 | 安徽科大讯飞信息科技股份有限公司 | Method and system for optimizing deep neural network |
| CN105025091A (en) * | 2015-06-26 | 2015-11-04 | 南京邮电大学 | Shop recommendation method based on position of mobile user |
| CN105183782A (en) * | 2015-08-14 | 2015-12-23 | 广东能龙教育股份有限公司 | Analysis recommendation method based on APP platform and data mining |
| CN105282616A (en) * | 2015-11-13 | 2016-01-27 | 云南大学 | Combined collaborative filtering IPTV program recommendation method |
| CN105430505A (en) * | 2015-11-13 | 2016-03-23 | 云南大学 | IPTV program recommending method based on combined strategy |
| CN105630742A (en) * | 2015-12-28 | 2016-06-01 | 广州酷狗计算机科技有限公司 | Feature vector calculation method and device |
| CN105786983A (en) * | 2016-02-15 | 2016-07-20 | 云南电网有限责任公司 | Employee individualized-learning recommendation method based on learning map and collaborative filtering |
| CN106530001A (en) * | 2016-11-03 | 2017-03-22 | 广州市万表科技股份有限公司 | Information recommending method and apparatus |
| CN106572185A (en) * | 2016-11-10 | 2017-04-19 | 北京邮电大学 | Method and apparatus for cooperative work of nodes in mobile Internet system |
| CN106897913A (en) * | 2017-01-22 | 2017-06-27 | 华南理工大学 | A kind of accurate injection machine selection method |
| CN106897430A (en) * | 2017-02-27 | 2017-06-27 | 温州职业技术学院 | The implementation method of digital library |
| CN106981228A (en) * | 2017-02-13 | 2017-07-25 | 上海大学 | A kind of interactive IT technical ability online education method |
| CN107194754A (en) * | 2017-04-11 | 2017-09-22 | 美林数据技术股份有限公司 | Stock trader's Products Show method based on mixing collaborative filtering |
| CN107220303A (en) * | 2017-05-10 | 2017-09-29 | 努比亚技术有限公司 | Recommendation method, device and the computer-readable medium of a kind of application |
| CN107329994A (en) * | 2017-06-08 | 2017-11-07 | 天津大学 | A kind of improvement collaborative filtering recommending method based on user characteristics |
| CN107665451A (en) * | 2016-07-29 | 2018-02-06 | 百度在线网络技术（北京）有限公司 | A kind of method and device that product push is carried out based on client characteristics data |
| CN107871276A (en) * | 2017-07-25 | 2018-04-03 | 平安科技（深圳）有限公司 | Inquiry unit, method and the computer-readable recording medium of loan product |
| CN108038629A (en) * | 2017-12-30 | 2018-05-15 | 北京工业大学 | A kind of optimization method based on collaborative filtering |
| CN108334592A (en) * | 2018-01-30 | 2018-07-27 | 南京邮电大学 | A kind of personalized recommendation method being combined with collaborative filtering based on content |
| CN108460145A (en) * | 2018-03-15 | 2018-08-28 | 北京邮电大学 | A kind of collaborative filtering recommending method based on mixing Interest Similarity |
| CN108804683A (en) * | 2018-06-13 | 2018-11-13 | 重庆理工大学 | Associate(d) matrix decomposes and the film of collaborative filtering recommends method |
| CN109214882A (en) * | 2018-07-09 | 2019-01-15 | 西北大学 | A kind of Method of Commodity Recommendation |
| CN109241422A (en) * | 2018-08-29 | 2019-01-18 | 广东工业大学 | A kind of improved user interest degree prediction technique based on collaborative filtering |
| CN109389447A (en) * | 2017-08-04 | 2019-02-26 | 北京京东尚科信息技术有限公司 | Item recommendation method, item recommendation system and computer-readable medium |
| CN109446430A (en) * | 2018-11-29 | 2019-03-08 | 西安电子科技大学 | Method, apparatus, computer equipment and the readable storage medium storing program for executing of Products Show |
| CN109478190A (en) * | 2016-08-29 | 2019-03-15 | 华为技术有限公司 | The region of search of recommender system reduces |
| CN109492154A (en) * | 2018-10-19 | 2019-03-19 | 广州市西美信息科技有限公司 | A kind of method of collaborative filtering recommending information |
| CN109509051A (en) * | 2018-09-12 | 2019-03-22 | 北京奇艺世纪科技有限公司 | A kind of item recommendation method and device |
| CN110059222A (en) * | 2019-04-24 | 2019-07-26 | 中山大学 | A kind of video tab adding method based on collaborative filtering |
| CN110210691A (en) * | 2018-04-27 | 2019-09-06 | 腾讯科技（深圳）有限公司 | Resource recommendation method, device, storage medium and equipment |
| CN110413880A (en) * | 2019-07-10 | 2019-11-05 | 合肥工业大学 | Based on user's personality single classification collaborative filtering method layered |
| CN111026966A (en) * | 2019-12-06 | 2020-04-17 | 创新奇智(成都)科技有限公司 | Search recommendation ranking method based on user, product portrait and correlation degree of user and product portrait |
| CN111144936A (en) * | 2019-12-19 | 2020-05-12 | 有米科技股份有限公司 | Similar population expansion method and device based on user tags |
| CN111160859A (en) * | 2019-12-26 | 2020-05-15 | 淮阴工学院 | Human resource post recommendation method based on SVD + + and collaborative filtering |
| CN111209469A (en) * | 2018-11-16 | 2020-05-29 | 海尔数字科技（上海）有限公司 | Personalized recommendation method and device, computer equipment and storage medium |
| CN111241360A (en) * | 2020-01-09 | 2020-06-05 | 腾讯科技（深圳）有限公司 | Information recommendation method, device, equipment and storage medium |
| CN112200601A (en) * | 2020-09-11 | 2021-01-08 | 深圳市法本信息技术股份有限公司 | Item recommendation method and device and readable storage medium |
| CN112967102A (en) * | 2021-02-04 | 2021-06-15 | 江苏警官学院 | Method for establishing customer portrait by logistics data |
| CN113821732A (en) * | 2021-11-24 | 2021-12-21 | 阿里巴巴达摩院(杭州)科技有限公司 | Item recommendation method and equipment for protecting user privacy and learning system |
| WO2022213662A1 (en) * | 2021-04-06 | 2022-10-13 | 深圳先进技术研究院 | Application recommendation method and system, terminal, and storage medium |
| CN117172796A (en) * | 2023-08-07 | 2023-12-05 | 北京智慧大王科技有限公司 | Big data electronic commerce management system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101458708A (en) * | 2008-12-05 | 2009-06-17 | 北京大学 | Searching result clustering method and device |
| CN101685458A (en) * | 2008-09-27 | 2010-03-31 | 华为技术有限公司 | Recommendation method and system based on collaborative filtering |
| CN102609523A (en) * | 2012-02-10 | 2012-07-25 | 上海视畅信息科技有限公司 | Collaborative filtering recommendation algorithm based on article sorting and user sorting |
-
2013
- 2013-01-16 CN CN201310016381.8A patent/CN103093376B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101685458A (en) * | 2008-09-27 | 2010-03-31 | 华为技术有限公司 | Recommendation method and system based on collaborative filtering |
| CN101458708A (en) * | 2008-12-05 | 2009-06-17 | 北京大学 | Searching result clustering method and device |
| CN102609523A (en) * | 2012-02-10 | 2012-07-25 | 上海视畅信息科技有限公司 | Collaborative filtering recommendation algorithm based on article sorting and user sorting |
Non-Patent Citations (1)
| Title |
|---|
| 黄章益等: "一种基于语义的中文文本特征降维技术研究", 《情报杂志》 * |
Cited By (66)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324690A (en) * | 2013-06-03 | 2013-09-25 | 焦点科技股份有限公司 | Mixed recommendation method based on factorization condition limitation Boltzmann machine |
| CN103279552A (en) * | 2013-06-06 | 2013-09-04 | 浙江大学 | Collaborative filtering recommendation method based on user interest groups |
| CN104636371A (en) * | 2013-11-11 | 2015-05-20 | 华为技术有限公司 | Information recommendation method and device |
| CN104636371B (en) * | 2013-11-11 | 2018-05-18 | 华为技术有限公司 | Information recommendation method and equipment |
| CN104751842A (en) * | 2013-12-31 | 2015-07-01 | 安徽科大讯飞信息科技股份有限公司 | Method and system for optimizing deep neural network |
| CN104751842B (en) * | 2013-12-31 | 2019-11-15 | 科大讯飞股份有限公司 | The optimization method and system of deep neural network |
| CN104111969A (en) * | 2014-06-04 | 2014-10-22 | 百度移信网络技术（北京）有限公司 | Method and system for measuring similarity |
| CN104111969B (en) * | 2014-06-04 | 2017-06-20 | 百度移信网络技术（北京）有限公司 | The method and system of a kind of similarity measurement |
| CN104731887A (en) * | 2015-03-13 | 2015-06-24 | 东南大学 | User similarity measuring method in collaborative filtering |
| CN104731887B (en) * | 2015-03-13 | 2018-02-02 | 东南大学 | A kind of user method for measuring similarity in collaborative filtering |
| CN105025091A (en) * | 2015-06-26 | 2015-11-04 | 南京邮电大学 | Shop recommendation method based on position of mobile user |
| CN105183782A (en) * | 2015-08-14 | 2015-12-23 | 广东能龙教育股份有限公司 | Analysis recommendation method based on APP platform and data mining |
| CN105430505A (en) * | 2015-11-13 | 2016-03-23 | 云南大学 | IPTV program recommending method based on combined strategy |
| CN105282616A (en) * | 2015-11-13 | 2016-01-27 | 云南大学 | Combined collaborative filtering IPTV program recommendation method |
| CN105430505B (en) * | 2015-11-13 | 2018-07-03 | 云南大学 | A kind of IPTV program commending methods based on combined strategy |
| CN105630742A (en) * | 2015-12-28 | 2016-06-01 | 广州酷狗计算机科技有限公司 | Feature vector calculation method and device |
| CN105630742B (en) * | 2015-12-28 | 2019-05-07 | 广州酷狗计算机科技有限公司 | Feature vector calculation method and device |
| CN105786983A (en) * | 2016-02-15 | 2016-07-20 | 云南电网有限责任公司 | Employee individualized-learning recommendation method based on learning map and collaborative filtering |
| CN107665451A (en) * | 2016-07-29 | 2018-02-06 | 百度在线网络技术（北京）有限公司 | A kind of method and device that product push is carried out based on client characteristics data |
| CN107665451B (en) * | 2016-07-29 | 2021-09-21 | 百度在线网络技术（北京）有限公司 | Method and device for pushing product based on customer characteristic data |
| CN109478190A (en) * | 2016-08-29 | 2019-03-15 | 华为技术有限公司 | The region of search of recommender system reduces |
| CN106530001A (en) * | 2016-11-03 | 2017-03-22 | 广州市万表科技股份有限公司 | Information recommending method and apparatus |
| CN106572185A (en) * | 2016-11-10 | 2017-04-19 | 北京邮电大学 | Method and apparatus for cooperative work of nodes in mobile Internet system |
| CN106897913A (en) * | 2017-01-22 | 2017-06-27 | 华南理工大学 | A kind of accurate injection machine selection method |
| CN106897913B (en) * | 2017-01-22 | 2020-10-27 | 华南理工大学 | Accurate type selection method for injection molding machine |
| CN106981228A (en) * | 2017-02-13 | 2017-07-25 | 上海大学 | A kind of interactive IT technical ability online education method |
| CN106897430A (en) * | 2017-02-27 | 2017-06-27 | 温州职业技术学院 | The implementation method of digital library |
| CN107194754A (en) * | 2017-04-11 | 2017-09-22 | 美林数据技术股份有限公司 | Stock trader's Products Show method based on mixing collaborative filtering |
| CN107220303A (en) * | 2017-05-10 | 2017-09-29 | 努比亚技术有限公司 | Recommendation method, device and the computer-readable medium of a kind of application |
| CN107329994A (en) * | 2017-06-08 | 2017-11-07 | 天津大学 | A kind of improvement collaborative filtering recommending method based on user characteristics |
| CN107871276A (en) * | 2017-07-25 | 2018-04-03 | 平安科技（深圳）有限公司 | Inquiry unit, method and the computer-readable recording medium of loan product |
| CN109389447A (en) * | 2017-08-04 | 2019-02-26 | 北京京东尚科信息技术有限公司 | Item recommendation method, item recommendation system and computer-readable medium |
| CN108038629A (en) * | 2017-12-30 | 2018-05-15 | 北京工业大学 | A kind of optimization method based on collaborative filtering |
| CN108334592A (en) * | 2018-01-30 | 2018-07-27 | 南京邮电大学 | A kind of personalized recommendation method being combined with collaborative filtering based on content |
| CN108334592B (en) * | 2018-01-30 | 2021-11-02 | 南京邮电大学 | Personalized recommendation method based on combination of content and collaborative filtering |
| CN108460145B (en) * | 2018-03-15 | 2020-07-03 | 北京邮电大学 | Collaborative filtering recommendation method based on mixed interest similarity |
| CN108460145A (en) * | 2018-03-15 | 2018-08-28 | 北京邮电大学 | A kind of collaborative filtering recommending method based on mixing Interest Similarity |
| CN110210691B (en) * | 2018-04-27 | 2024-02-06 | 腾讯科技（深圳）有限公司 | Resource recommendation method, device, storage medium and equipment |
| CN110210691A (en) * | 2018-04-27 | 2019-09-06 | 腾讯科技（深圳）有限公司 | Resource recommendation method, device, storage medium and equipment |
| CN108804683B (en) * | 2018-06-13 | 2021-11-23 | 重庆理工大学 | Movie recommendation method combining matrix decomposition and collaborative filtering algorithm |
| CN108804683A (en) * | 2018-06-13 | 2018-11-13 | 重庆理工大学 | Associate(d) matrix decomposes and the film of collaborative filtering recommends method |
| CN109214882B (en) * | 2018-07-09 | 2021-06-25 | 西北大学 | Commodity recommendation method |
| CN109214882A (en) * | 2018-07-09 | 2019-01-15 | 西北大学 | A kind of Method of Commodity Recommendation |
| CN109241422A (en) * | 2018-08-29 | 2019-01-18 | 广东工业大学 | A kind of improved user interest degree prediction technique based on collaborative filtering |
| CN109509051A (en) * | 2018-09-12 | 2019-03-22 | 北京奇艺世纪科技有限公司 | A kind of item recommendation method and device |
| CN109492154A (en) * | 2018-10-19 | 2019-03-19 | 广州市西美信息科技有限公司 | A kind of method of collaborative filtering recommending information |
| CN111209469A (en) * | 2018-11-16 | 2020-05-29 | 海尔数字科技（上海）有限公司 | Personalized recommendation method and device, computer equipment and storage medium |
| CN109446430B (en) * | 2018-11-29 | 2021-10-01 | 西安电子科技大学 | Product recommendation method and device, computer equipment and readable storage medium |
| CN109446430A (en) * | 2018-11-29 | 2019-03-08 | 西安电子科技大学 | Method, apparatus, computer equipment and the readable storage medium storing program for executing of Products Show |
| CN110059222A (en) * | 2019-04-24 | 2019-07-26 | 中山大学 | A kind of video tab adding method based on collaborative filtering |
| CN110413880A (en) * | 2019-07-10 | 2019-11-05 | 合肥工业大学 | Based on user's personality single classification collaborative filtering method layered |
| CN110413880B (en) * | 2019-07-10 | 2021-09-14 | 合肥工业大学 | Single-classification collaborative filtering method based on user personality hierarchical structure |
| CN111026966B (en) * | 2019-12-06 | 2023-12-22 | 创新奇智(成都)科技有限公司 | Search recommendation ordering method based on user and product portrait and association degree of user and product portrait |
| CN111026966A (en) * | 2019-12-06 | 2020-04-17 | 创新奇智(成都)科技有限公司 | Search recommendation ranking method based on user, product portrait and correlation degree of user and product portrait |
| CN111144936A (en) * | 2019-12-19 | 2020-05-12 | 有米科技股份有限公司 | Similar population expansion method and device based on user tags |
| CN111144936B (en) * | 2019-12-19 | 2024-01-05 | 有米科技股份有限公司 | Similar crowd expansion method and device based on user labels |
| CN111160859A (en) * | 2019-12-26 | 2020-05-15 | 淮阴工学院 | Human resource post recommendation method based on SVD + + and collaborative filtering |
| CN111241360A (en) * | 2020-01-09 | 2020-06-05 | 腾讯科技（深圳）有限公司 | Information recommendation method, device, equipment and storage medium |
| CN111241360B (en) * | 2020-01-09 | 2023-03-21 | 深圳市雅阅科技有限公司 | Information recommendation method, device, equipment and storage medium |
| CN112200601B (en) * | 2020-09-11 | 2024-05-14 | 深圳市法本信息技术股份有限公司 | Item recommendation method, device and readable storage medium |
| CN112200601A (en) * | 2020-09-11 | 2021-01-08 | 深圳市法本信息技术股份有限公司 | Item recommendation method and device and readable storage medium |
| CN112967102A (en) * | 2021-02-04 | 2021-06-15 | 江苏警官学院 | Method for establishing customer portrait by logistics data |
| WO2022213662A1 (en) * | 2021-04-06 | 2022-10-13 | 深圳先进技术研究院 | Application recommendation method and system, terminal, and storage medium |
| CN113821732B (en) * | 2021-11-24 | 2022-02-18 | 阿里巴巴达摩院(杭州)科技有限公司 | Item recommendation method and equipment for protecting user privacy and learning system |
| CN113821732A (en) * | 2021-11-24 | 2021-12-21 | 阿里巴巴达摩院(杭州)科技有限公司 | Item recommendation method and equipment for protecting user privacy and learning system |
| CN117172796A (en) * | 2023-08-07 | 2023-12-05 | 北京智慧大王科技有限公司 | Big data electronic commerce management system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103093376B (en) | 2020-02-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103093376A (en) | Clustering collaborative filtering recommendation system based on singular value decomposition algorithm | |
| CN104063481B (en) | A kind of film personalized recommendation method based on the real-time interest vector of user | |
| CN108205768B (en) | Database establishing method, data recommending device, equipment and storage medium | |
| CN110222272A (en) | A kind of potential customers excavate and recommended method | |
| Ba et al. | Clustering collaborative filtering recommendation system based on SVD algorithm | |
| CN107909433A (en) | A kind of Method of Commodity Recommendation based on big data mobile e-business | |
| US20230306491A1 (en) | User feature-based page displaying method and apparatus, medium, and electronic device | |
| CN106548255A (en) | A kind of Method of Commodity Recommendation based on mass users behavior | |
| CN102968506A (en) | Personalized collaborative filtering recommendation method based on extension characteristic vectors | |
| James et al. | Penalized and constrained optimization: an application to high-dimensional website advertising | |
| CN104463637A (en) | Commodity recommendation method and device based on electronic business platform and server | |
| CN103778214A (en) | Commodity property clustering method based on user comments | |
| US20220122100A1 (en) | Product evaluation system and method of use | |
| CN110503497A (en) | A kind of electric business management platform Method of Commodity Recommendation of the consumption habit based on client | |
| CN105809558A (en) | Social network based recommendation method and apparatus | |
| CN106022869A (en) | Consumption object recommending method and consumption object recommending device | |
| CN108198045A (en) | The design method of mixing commending system based on e-commerce website data mining | |
| Jia et al. | Users' brands preference based on SVD++ in recommender systems | |
| CN108573041A (en) | Probability matrix based on weighting trusting relationship decomposes recommendation method | |
| CN103886486A (en) | Electronic commerce recommending method based on support vector machine (SVM) | |
| CN106447464A (en) | Electronic commerce data processing system | |
| CN108122153A (en) | Personalized recommendation method based on cloud computing tupe under e-commerce environment | |
| CN105931082A (en) | Commodity category keyword extraction method and device | |
| CN108109030A (en) | Data analysis method, system and non-transient computer readable medium | |
| CN110599281A (en) | Method and device for determining target shop |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |