CN104156388A - Collaborative filtering recommendation method based on trustful privacy maintenance in personalized search - Google Patents
Collaborative filtering recommendation method based on trustful privacy maintenance in personalized search Download PDFInfo
- Publication number
- CN104156388A CN104156388A CN201410308536.XA CN201410308536A CN104156388A CN 104156388 A CN104156388 A CN 104156388A CN 201410308536 A CN201410308536 A CN 201410308536A CN 104156388 A CN104156388 A CN 104156388A
- Authority
- CN
- China
- Prior art keywords
- user
- collaborative filtering
- targeted customer
- degree
- filtering 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.)
- 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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
- G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioethics (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a collaborative filtering recommendation method based on trustful privacy maintenance in personalized search and belongs to the field of computer networks. According to the basic idea, firstly, the trust degree Tcf of a target user to the execution behavior of a collaborative filtering recommendation system is established by means of automatic trust negotiation between the target user and the collaborative filtering recommendation system; the target user judges the degree of disclosure of personal information according to the trust degree Tcf, the collaborative filtering recommendation system collects the personal information provided by the target user and conducts clustering on users to obtain an initial user set with the interest similar to that of the target user; finally, the target user selects a similar user with the trust degree not lower than the Tcf from the user set and makes the similar user as a recommended object. The collaborative filtering recommendation method based on trustful privacy maintenance can overcome the defect that an existing collaborative filtering recommendation method is short of privacy protection in the search and modeling process of personal information, control over the degree of disclosure of the personal information by the target user is achieved, and comprehensive protection is provided for privacy data of the users.
Description
Technical field
The present invention relates to the collaborative filtering recommending method keeping based on the privacy of trusting in personalized search, basic thought is: first between targeted customer and Collaborative Filtering Recommendation System, pass through automated trust negotiation, set up the degree of belief T of the act of execution of targeted customer to Collaborative Filtering Recommendation System
cf; Targeted customer will be according to degree of belief T
cfjudge the degree of leakage of personal information, the personal information that Collaborative Filtering Recommendation System collection targeted customer provides also obtains an initial user set similar to targeted customer's interest to user clustering; Ideal user therefrom selects degree of belief to be not less than T
cfsimilar users as recommended.The collaborative filtering recommending method keeping based on the privacy of trusting can make up the defect that existing collaborative filtering recommending method lacks secret protection in the collection modeling process of personal information; realize the control of the disclosure degree of targeted customer to self personal information, for privacy of user data provide comprehensive protection.Belong to computer network field.
Background technology
Web has become an important channel of people's obtaining information, growing due to Web information, and people have to spend a large amount of time removal search, browse the information of oneself needs.Search engine (search engine) is the instrument of the most general auxiliary people's retrieving information, such as traditional search engine AltaVista, and Yahoo and search engine Google of new generation etc.Information retrieval technique has met the certain needs of people, but due to its general character, still can not meet the inquiry request of different background, different object and different times.Commending system based on collaborative filtering puts forward for this problem, and it provides different services for different user, to meet the different needs.Collaborative filtering is to recommend resource according to user's similarity.It compares the described file of user, and by user clustering, recommends resource according to similar users, so can recommend the interested content making new advances for user, realizes the object of initiatively recommending.Collaborative filtering recommending technology is one of the most successful in personalized service development and most widely used recommended technology.
But in the behind of this slice prosperity, a huge threat starts little by little to appear in one's mind out.Developing rapidly of internet increases global netizen's number continuing detonation formula, the up-to-date report demonstration that the Forrester Research of research institution announces, and before 2013, global netizen's number is expected to realize more than 45% high growth.Expect global netizen's number in 2013 and will reach 2,200,000,000, wherein Asia netizen's ratio will be up to 43%, becomes topmost growth area.In this numerous Internet user, having many people all can suffer from various types of privacies every day invades.People unavoidably start worry in the time of search service easy to use, and whether the individual privacy of oneself is also in safety area actually.Because the commending system based on collaborative filtering must be collected user's personal information, thereby can provide the Search Results that meets its personal interest preference for different user.Relate in this course that collection modeling to userspersonal information, transmission are used and the operation such as access, inevitably caused the safety problem of privacy of user.The information leakage that for example Internet Transmission causes, meets with the data that cause of assault and reveals, the access control policy data leakage that causes etc. of slipping up.Individual privacy often comprises the information with important value, if these information are obtained by other people, likely can cause personal reputation loss, emotional distress or economic loss, and in view of this privacy becomes the secret information of individual's hope.And what is more important, in these and user-dependent information, except individual privacy, also likely relate to the confidential information of national defence and government, national security is caused to immeasurable threat, for example, reveal direction, military technology progress of research of the current strategic research of China etc.
Consider for privacy, many users are unwilling to provide personal information, and this has greatly hindered the universal and application of Collaborative Filtering Recommendation System.Therefore, how to allow user provide the necessary information that commending system uses and the individual privacy of not revealing user to become subject matter in the urgent need to address in Collaborative Filtering Recommendation System development.The existing commending system based on collaborative filtering is based on there being similar hobby between user, calculates similarity therebetween, for targeted customer chooses the similar neighbours of interest, and is targeted customer's recommended products by it.The life cycle of privacy of user in Collaborative Filtering Recommendation System comprises three phases: search modeling, transmission are used and access access.The search modelling phase is considered to heavy in the crowd of secret protection scheme, and at this one-phase, thereby need to collect user's personal information and by modeling, user clustering be found to similar users be that targeted customer recommends the interested content making new advances to commending system.If just there is the leakage of personal information and privacy in this link, even if the security of other links also cannot fundamentally ensure the safety of privacy information at height so.
Current present Research shows that targeted customer always supposes that used collaborative filtering system is believable; but being existing Collaborative Filtering Recommendation System, the fact lacks the protection to personal information and privacy; commending system explicitly or not explicitly are informed to user; in search procedure, personal information will be collected and be utilized; but the uncontrollable commending system of user is collected which kind of information of oneself and collected which kind of degree, and by how to choose the similar neighbours of interest by modeling comes for its recommended products.Therefore user's individual privacy cannot obtain sufficient safety guarantee in userspersonal information's collection modeling process.Especially when targeted customer and commending system be not during at same security domain, conventional access control method (as MAC, RBAC etc.) can not be controlled both behaviors effectively.
The collaborative filtering recommending method for this reason keeping in the urgent need to designing a novel privacy, provides personalized secret protection means in the collection modelling phase of personal information.
Summary of the invention
The object of the present invention is to provide the collaborative filtering recommending method keeping based on the privacy of trusting in a kind of personalized search.The method is first by the mode of the automated trust negotiation relation that breaks the wall of mistrust between targeted customer and Collaborative Filtering Recommendation System, and after automated trust negotiation finishes, targeted customer obtains the degree of belief T of the act of execution to Collaborative Filtering Recommendation System
cf; Then targeted customer is according to T
cfjudge the degree of leakage of personal information, the personal information that Collaborative Filtering Recommendation System collection targeted customer provides also obtains some users similar to targeted customer's interest to user clustering; Ideal user therefrom selects degree of belief to be not less than T
cfsimilar users as recommended.Advantage of the present invention is that the collaborative filtering recommending mode keeping based on the privacy of trusting can be set up the trust between targeted customer and Collaborative Filtering Recommendation System by the mode of automated trust negotiation; and instruct targeted customer that the level of detail of personal information is provided according to the degree of belief of setting up in automated trust negotiation process, make up the defect that existing collaborative filtering recommending method lacks secret protection in the collection modeling process of personal information.Mode based on trusting in the selection course of this external recommended is screened recommended, greatly reduced malice, insecure similar users selecteed chance.
For achieving the above object, the present invention takes following technical scheme:
The present invention comprises the establishment stage of trusting relationship between targeted customer and Collaborative Filtering Recommendation System while specifically enforcement; User clustering stage and recommended choice phase.First by automated trust negotiation, obtain the degree of belief T of targeted customer to Collaborative Filtering Recommendation System
cf.Wherein targeted customer and Collaborative Filtering Recommendation System are deployed respectively automated trust negotiation administration module, as Fig. 1.Automated trust negotiation administration module is responsible for by the exchange of the attribute information between targeted customer and the Collaborative Filtering Recommendation System relation that breaks the wall of mistrust.Automated trust negotiation administration module comprises again two parts: consultative management device and attribute information storehouse.Attribute information library storage participant's certificate and access control policy information.Consultative management device is responsible for submitting relevant certificate to according to the other side's negotiation request.In the process of trust negotiation, need certificate exchange repeatedly, until set up trusting relationship.After automated trust negotiation finishes, targeted customer obtains the degree of belief T of the act of execution to Collaborative Filtering Recommendation System
cf.Targeted customer is according to degree of belief T
cfjudge the degree of leakage of personal information, the personal information that Collaborative Filtering Recommendation System collection targeted customer provides is also obtained an initial user set similar to targeted customer's interest and similar users set is submitted to targeted customer user clustering by the mode of calculating user's similarity.Targeted customer therefrom selects degree of belief to be not less than T
cfsimilar users as recommended.Specific practice is to evaluate user's recommendation behavior by degree of belief, and instructs the selection of recommended with it.Each user is after being selected as recommended, and user can assess this recommendation behavior each other, and the result obtaining is called local trust degree.Local trust degree, as raw data, is inputted into trust model.User's overall confidence level, by occurring with it to recommend other users of behavior local trust degree to it, and these users' overall confidence level is calculated.Each user can upgrade its local trust degree table after completing recommendation.The user that this mode can reduce malice is as much as possible selected as the chance of recommended.Last selecteed recommended is responsible for user and recommends the interested content making new advances.
The collaborative filtering recommending method keeping based on the privacy of trusting that the present invention proposes compared with prior art, has following significantly advantage and beneficial effect:
1) instruct targeted customer that the level of detail of personal information is provided according to the trusting relationship of setting up in automated trust negotiation process, realized targeted customer to privacy degree of exposure from main control.
2) mode keeping based on the privacy of trusting, has made up the defect that existing collaborative filtering recommending method lacks secret protection in the collection modelling phase of personal information.
3) in recommended choice phase process integrated trust management technology can reduce malice, the selecteed chance of insecure similar users, be excluded in recommended select outside.
The collaborative filtering recommending method keeping based on the privacy of trusting that the present invention proposes, can comprehensively protect user's private data in the collection modelling phase of personal information, and be that it recommends new interested content for targeted customer provides reliable recommended.The method is intended to provide a kind of collaborative filtering recommending mechanism that personalized secret protection is provided, and finally solves the defect problem that existing collaborative filtering recommending method lacks secret protection in the collection modeling process of personal information.
Brief description of the drawings
Fig. 1 automated trust negotiation administration module;
Fig. 2 specific implementation process;
Embodiment
Specific implementation process, as Fig. 2, has following characteristics:
Whole personalized search comprises Collaborative Filtering Recommendation System, targeted customer, similar users, recommended and trust data.
The present invention comprises the establishment stage of trusting relationship between targeted customer and Collaborative Filtering Recommendation System while specifically enforcement; User clustering stage and recommended choice phase.
The establishment stage of trusting relationship between targeted customer and Collaborative Filtering Recommendation System:
First between targeted customer and Collaborative Filtering Recommendation System by the mode of the automated trust negotiation relation that breaks the wall of mistrust.In the process of trust negotiation, need certificate exchange repeatedly, until set up trusting relationship.After automated trust negotiation finishes, targeted customer obtains the degree of belief T of the act of execution to Collaborative Filtering Recommendation System
cf.
The user clustering stage:
Targeted customer is according to degree of belief T
cfjudge the degree of leakage of personal information, the personal information that Collaborative Filtering Recommendation System collection targeted customer provides also adopts Person degree of correlation formula to calculate user similarity sim (i, j), represents as follows:
Wherein R
iand R
jbe respectively the average score of user i and j, R
i, cwith R
j, cfor user i and j are to jointly commenting undue project c scoring, m is the number of user i and the common scoring of j.
Collaborative Filtering Recommendation System is by calculating user's similarity sim (i, j) mode obtains initial user's set similar to targeted customer's interest after to user clustering, and then targeted customer is submitted in similar users set by Collaborative Filtering Recommendation System.
The recommended choice phase:
Targeted customer selects degree of belief to be not less than T from similar users set
cfsimilar users as recommended.Specific practice is to evaluate user's recommendation behavior by degree of belief, and instructs the selection of recommended with it.Each user is after being selected as recommended, and user can assess this recommendation behavior each other, and the result obtaining is called local trust degree.Local trust degree, as raw data, is inputted into trust model.User's overall confidence level, by occurring with it to recommend other users of behavior local trust degree to it, and these users' overall confidence level is calculated.Each user can upgrade its local trust degree table after completing recommendation.Only have global reputation to be not less than T
cfsimilar users to be just selected as recommended be that targeted customer recommends the interested content making new advances.The user that this mode can reduce malice is as much as possible selected as the chance of recommended.The overall confidence level computing formula of similar users is as follows:
Wherein V
sthe overall confidence level of representative of consumer x, S is user's set that behavior occurred to recommend with user x.F
yxthe local confidence value of user y to user x.W
ylocal confidence level f
yxweight.The computation process of whole trust adopts the method repeatedly iterating, until V
xconverge to a stable value.
Finally it should be noted that: above embodiment is only in order to illustrate the present invention and unrestricted technical scheme described in the invention; Therefore, although this instructions has been described in detail the present invention with reference to each above-mentioned embodiment,, those of ordinary skill in the art should be appreciated that still and can modify or be equal to replacement the present invention; And all do not depart from technical scheme and the improvement thereof of the spirit and scope of invention, it all should be encompassed in the middle of claim scope of the present invention.
Claims (4)
1. the collaborative filtering recommending method keeping based on the privacy of trusting in a personalized search, instruct targeted customer that the level of detail of personal information is provided according to the trusting relationship of setting up in automated trust negotiation process, realized targeted customer to privacy degree of exposure from main control; In recommended choice phase process integrated trust management technology can reduce malice, the selecteed chance of insecure similar users, be excluded in recommended select outside; It is characterized in that comprising the following steps:
The establishment stage of trusting relationship between targeted customer and Collaborative Filtering Recommendation System:
Between targeted customer and Collaborative Filtering Recommendation System, by the mode of the automated trust negotiation relation that breaks the wall of mistrust, after automated trust negotiation finishes, targeted customer obtains the degree of belief T of the act of execution to Collaborative Filtering Recommendation System
cf;
The user clustering stage:
Targeted customer is according to degree of belief T
cfjudge the degree of leakage of personal information, the personal information that Collaborative Filtering Recommendation System collection targeted customer provides also adopts Person degree of correlation formula to calculate user's similarity sim (i, j), to obtaining an initial user set similar to targeted customer's interest after user clustering, and similar users set is submitted to targeted customer;
The recommended choice phase:
Targeted customer selects global reputation to be not less than T from similar users set
cfsimilar users as recommended.
2. the collaborative filtering recommending method keeping based on the privacy of trusting in personalized search according to claim 1, it is characterized in that: targeted customer and Collaborative Filtering Recommendation System are deployed respectively automated trust negotiation administration module, automated trust negotiation administration module is responsible for by the exchange of the attribute information between targeted customer and the Collaborative Filtering Recommendation System relation that breaks the wall of mistrust; Automated trust negotiation administration module comprises again two parts: consultative management device and attribute information storehouse, attribute information library storage participant's certificate and access control policy information, consultative management device is responsible for submitting relevant certificate to according to the other side's negotiation request, after automated trust negotiation finishes, targeted customer obtains the degree of belief T of the act of execution to Collaborative Filtering Recommendation System
cf.
3. the collaborative filtering recommending method keeping based on the privacy of trusting in personalized search according to claim 1, is characterized in that: targeted customer is according to degree of belief T
cfjudge the degree of leakage of personal information, the mode that the personal information that Collaborative Filtering Recommendation System is collected targeted customer to be provided is also passed through to calculate user's similarity is to user clustering, adopt Person degree of correlation formula to calculate user similarity sim (i, j), formulae express is as follows:
Wherein R
iand R
jbe respectively the average score of user i and j, R
i, cwith R
j, cfor user i and j are to jointly commenting undue project c scoring, m is the number of user i and the common scoring of j.
4. the collaborative filtering recommending method keeping based on the privacy of trusting in personalized search according to claim 1, is characterized in that: in the process of code transmission, adopt faith mechanism evaluation user's recommendation behavior, only have global reputation to be not less than T
cfsimilar users just can be selected as recommended; The overall confidence level of similar users, by occurring with it to recommend other users of behavior local trust degree to it, and these users' overall confidence level calculates, and can adopt following formula to explain:
Wherein V
xthe overall confidence level of representative of consumer x, S is user's set that behavior occurred to recommend with user x.F
yxthe local confidence value of user y to user x.W
ylocal confidence level f
yxweight.The computation process of whole trust adopts the method repeatedly iterating, until V
xconverge to a stable value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410308536.XA CN104156388A (en) | 2014-06-26 | 2014-06-26 | Collaborative filtering recommendation method based on trustful privacy maintenance in personalized search |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410308536.XA CN104156388A (en) | 2014-06-26 | 2014-06-26 | Collaborative filtering recommendation method based on trustful privacy maintenance in personalized search |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104156388A true CN104156388A (en) | 2014-11-19 |
Family
ID=51881892
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410308536.XA Pending CN104156388A (en) | 2014-06-26 | 2014-06-26 | Collaborative filtering recommendation method based on trustful privacy maintenance in personalized search |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104156388A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105427050A (en) * | 2015-12-02 | 2016-03-23 | 常州大学 | Trust model based food quality evaluation method |
CN105574430A (en) * | 2015-12-02 | 2016-05-11 | 西安邮电大学 | Novel privacy protection method in collaborative filtering recommendation system |
CN105653640A (en) * | 2015-12-25 | 2016-06-08 | 江苏东大金智信息系统有限公司 | Collaborative filtering recommendation method based on trust mechanism |
WO2016149929A1 (en) * | 2015-03-26 | 2016-09-29 | Nokia Technologies Oy | Method, apparatus and computer program product for identifying a vulnerable friend for privacy protection in a social network |
CN107491557A (en) * | 2017-09-06 | 2017-12-19 | 徐州医科大学 | A kind of TopN collaborative filtering recommending methods based on difference privacy |
CN111259260A (en) * | 2020-03-30 | 2020-06-09 | 九江学院 | Privacy protection method in personalized recommendation based on sorting classification |
CN112232367A (en) * | 2020-09-10 | 2021-01-15 | 山东师范大学 | Network behavior similarity judgment method and system |
WO2022116422A1 (en) * | 2020-12-01 | 2022-06-09 | 平安科技(深圳)有限公司 | Product recommendation method and apparatus, and electronic device and computer-readable storage medium |
-
2014
- 2014-06-26 CN CN201410308536.XA patent/CN104156388A/en active Pending
Non-Patent Citations (2)
Title |
---|
廖振松 等: "自动信任协商及其发展趋势", 《软件学报》 * |
李涛 等: "一种基于用户聚类的协同过滤推荐算法", 《系统工程与电子技术》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016149929A1 (en) * | 2015-03-26 | 2016-09-29 | Nokia Technologies Oy | Method, apparatus and computer program product for identifying a vulnerable friend for privacy protection in a social network |
CN105427050A (en) * | 2015-12-02 | 2016-03-23 | 常州大学 | Trust model based food quality evaluation method |
CN105574430A (en) * | 2015-12-02 | 2016-05-11 | 西安邮电大学 | Novel privacy protection method in collaborative filtering recommendation system |
CN105574430B (en) * | 2015-12-02 | 2018-04-06 | 西安邮电大学 | A kind of new method for secret protection in Collaborative Filtering Recommendation System |
CN105653640A (en) * | 2015-12-25 | 2016-06-08 | 江苏东大金智信息系统有限公司 | Collaborative filtering recommendation method based on trust mechanism |
CN107491557A (en) * | 2017-09-06 | 2017-12-19 | 徐州医科大学 | A kind of TopN collaborative filtering recommending methods based on difference privacy |
CN111259260A (en) * | 2020-03-30 | 2020-06-09 | 九江学院 | Privacy protection method in personalized recommendation based on sorting classification |
CN111259260B (en) * | 2020-03-30 | 2023-06-02 | 九江学院 | Privacy protection method in personalized recommendation based on sorting classification |
CN112232367A (en) * | 2020-09-10 | 2021-01-15 | 山东师范大学 | Network behavior similarity judgment method and system |
CN112232367B (en) * | 2020-09-10 | 2022-06-21 | 山东师范大学 | Network behavior similarity judgment method and system |
WO2022116422A1 (en) * | 2020-12-01 | 2022-06-09 | 平安科技(深圳)有限公司 | Product recommendation method and apparatus, and electronic device and computer-readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104156388A (en) | Collaborative filtering recommendation method based on trustful privacy maintenance in personalized search | |
US10721268B2 (en) | Systems and user interfaces for dynamic and interactive investigation based on automatic clustering of related data in various data structures | |
Xu et al. | Link prediction with signed latent factors in signed social networks | |
CA2805391C (en) | Determining relevant information for domains of interest | |
Ling et al. | Selection of model in developing information security criteria on smart grid security system | |
CN105447113B (en) | A kind of information analysis method based on big data | |
Zhang et al. | Detection of shilling attacks in recommender systems via spectral clustering | |
TW201901539A (en) | Assessment method of risk, device, computer device and storage medium | |
JP2022519208A (en) | Real-time event detection on a social data stream | |
CN107895038A (en) | A kind of link prediction relation recommends method and device | |
CN104199875A (en) | Search recommending method and device | |
CN107862219A (en) | The guard method of demand privacy in a kind of social networks | |
CN104301323B (en) | Balanced third-party application personalized service and the method for user privacy information safety | |
Zhang et al. | Reverse attack: Black-box attacks on collaborative recommendation | |
CN104182518A (en) | Collaborative filtering recommendation method and device | |
De et al. | Differentially private link prediction with protected connections | |
Arazzi et al. | Turning privacy-preserving mechanisms against federated learning | |
Ahmed et al. | Aggregation of security metrics for decision making: a reference architecture | |
CN109614521A (en) | A kind of efficient secret protection subgraph inquiry processing method | |
Xu et al. | Leakage-abuse attacks against forward and backward private searchable symmetric encryption | |
CN106933880A (en) | A kind of label data leaks channel detection method and device | |
CN106611339B (en) | Seed user screening method, and product user influence evaluation method and device | |
CN106506526B (en) | A kind of verification method and system using recognition result | |
Pan et al. | CGStream: continuous correlated graph query for data streams | |
CN115718927A (en) | Difference privacy mixed recommendation method based on untrusted server |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141119 |
|
WD01 | Invention patent application deemed withdrawn after publication |