CN112381112B - User identity recognition method and system based on multi-mode item set of user data - Google Patents

User identity recognition method and system based on multi-mode item set of user data Download PDF

Info

Publication number
CN112381112B
CN112381112B CN202011109382.3A CN202011109382A CN112381112B CN 112381112 B CN112381112 B CN 112381112B CN 202011109382 A CN202011109382 A CN 202011109382A CN 112381112 B CN112381112 B CN 112381112B
Authority
CN
China
Prior art keywords
similarity
user
decision
item set
extensible
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.)
Active
Application number
CN202011109382.3A
Other languages
Chinese (zh)
Other versions
CN112381112A (en
Inventor
杨灿
王澜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN202011109382.3A priority Critical patent/CN112381112B/en
Publication of CN112381112A publication Critical patent/CN112381112A/en
Application granted granted Critical
Publication of CN112381112B publication Critical patent/CN112381112B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a user identity recognition method and a system of a multimode item set based on user data, wherein the method comprises the following steps: respectively constructing extensible multimode item sets of historical data and user data to be identified; carrying out similarity measurement on the extensible multi-mode item set of the historical data and the extensible multi-mode item set of the user data to be identified; and making a decision on the user identification result according to the similarity measurement result. The method and the device comprehensively improve the performance of the user identification method from the aspects of constructing a multi-mode item set, measuring the similarity and fusing decisions, and can be widely applied to the fields of personalized recommendation, information evidence obtaining and the like.

Description

User identity recognition method and system based on multi-mode item set of user data
Technical Field
The present application relates to the field of user identification, and in particular, to a user identification method and system for a multimode item set based on user data.
Background
Currently, with the vigorous development of the internet and big data technology, people can enjoy the convenience brought by the internet and big data technology, and meanwhile, a plurality of digital marks are left carelessly, so that effective data is provided for the research of the user identification problem. The statistical feature matching by using the digital trace left by people is an important method for solving the problem of user identification. However, how to achieve accurate user identification using limited digital features is a matter of controversy for researchers.
The trace of activity left by a person in the information space naturally forms a sequence over time, e.g. a website visited by the user, a purchased merchandise, a channel watched over time forms a sequence reflecting the corresponding behavior characteristics of the user. We focus on the more frequent user behavior (also called frequent items) in the sequence, and further study is done with the collection of frequent item sets thus made. Because frequent item sets can stably maintain their association with the user behavior subject over a longer time frame and are difficult to hide. Accordingly, there is an urgent need in the industry to develop a method or system for implementing user identification that exploits this feature to build an extensible multimodal set.
Disclosure of Invention
The application mainly aims to provide a user identity recognition method and system based on a multi-mode item set of user data, which can more effectively improve the performance of the user identity recognition method by constructing the multi-mode item set, similarity measurement and fusion decision.
In order to achieve the above purpose, the present application proposes the following technical scheme:
a user identity recognition method based on a multi-mode item set of user data comprises the following steps:
constructing an expandable multi-mode item set of user data to be identified;
performing similarity measurement on the extensible multi-mode item set of the user data to be identified and the extensible multi-mode item set of the pre-constructed historical data;
and making a decision on the user identification result according to the similarity measurement result to obtain the identification result.
Preferably, the step of constructing the extensible multi-modal set of user data to be identified and the step of constructing the extensible multi-modal set of history data are both MISFUB, the MISFUB specifically comprising:
s2.1, constructing an extensible multimode item; the behavior, state or event of the user is denoted as Item, and the Item Sequence of the user is denoted as Item sequence= { Item 1 ,Item 2 ,Item 3 ,……,Item Nr -where Nr is a positive integer;
the modulus of the extensible item set is denoted as n; wherein the level 1 Item set is the Item Sequence in the natural state and is marked as C 1 ＝{S i I = 1,2, … …, nr }; level 2 item set is denoted as C 2 ＝{S ij I=1, 2, … …, nr-1 and j=i+1, … …, nr }; level 3 item set is denoted as C 3 ＝{S ijk I=1, 2, … …, nr-2 and j=i+1, … …, nr-1 and k=i+2, … …, nr }; n-level item set is denoted as C n ＝{S i1i2…in |i 1 =1, 2, … …, nr-n+1 and i k ＝i 1 +k-1,k＝1,2,……,n}；S i1i2…in Representing a sequence of n items from Nr items;
s2.2, summing the n multimode item sets obtained in the step S2.1, namely:
s2.3 pairIs ordered in descending order according to the frequency of occurrence of each sequence S, namely:wherein->Is->M=1, 2, … …, length +.>
S2.4, define the ratio r top The method comprises the following steps:
according to r top Extraction ofThe first sequences of the middle row form F n The method comprises the following steps:
preferably, the similarity measurement of the extensible multi-modal set of user data to be identified and the extensible multi-modal set of pre-structured historical data specifically comprises:
the ith user in the history data is noted as u xi Its corresponding extensible multi-modal set is denoted asThe j-th user in the user data to be identified is marked as u yj Its corresponding extensible multimode item set is denoted +.>The similarity measure is defined as:
this method is called JKL, wherein,represents u calculated by JKL xi And u yj Similarity of (2); />Representing u calculated using Jaccard coefficients xi And u yj The similarity of (2), the formula is:
wherein,representation->And->Length of intersection, i.e.)> Representation->And->Length of union, i.e.)>
Represents u calculated by KL method xi And u yj The similarity of (2), the formula is:
wherein,and->Respectively indicate->And->Frequency histogram of intersection, D (||·) is KL divergence function expressed as:
where k= |α= |β| represents the length of α or β.
Preferably, the user recognition result in making a decision on the user recognition result according to the similarity measurement result is defined as:
wherein V is y ＝{u yj I j e {1,2,.,... ij Is u xi And u yj Similarity of (i.e. w) ij ＝similarity(u xi ，u yj )；
The decision making of the user identification result comprises two decision schemes:
decision scheme based on a similarity measure: the identification result can be obtained by applying a similarity measurement method, and the formula is as follows:
wherein k represents a certain similarity measurement method, and Jaccard coefficients, KL divergence, cosine similarity and Pearson correlation coefficients can be selected;
decision schemes based on various similarity metrics: the scheme is called FI, is a fusion decision method for intersecting candidate sets generated based on various similarity measurement methods, and is defined as the formula:
wherein, K epsilon {1,2, & gt..A.K.. Represents the number of the similarity measurement method used, and the results obtained by the K similarity measurement methods are intersected to obtain the final recognition result.
Preferably, the decision schemes based on the multiple similarity measures include two types of fusion decision schemes:
(1) The similarity of the same model fuses the decision, the formula is defined as:
wherein K is {1,2,.. The number of the similarity measurement method used is denoted by K, n is a modulus of the expandable item set, and a positive integer is taken;
(2) A multi-class modulus similarity fusion decision, the formula is defined as:
where k.epsilon. {1, 2.. The number of similarity measures used is denoted by K, N e {1, 2.,. N represents the modulo taking of the extensible item set for N different cases.
Preferably, the accuracy calculation formula of the decision scheme based on the multiple similarity measures is defined as follows:
wherein V is x ＝{u xi I e {1,2,.,. Nu }, represents a set of Nu users in the history data;
because ofPossibly empty, the reject rate is defined as:
preferably, the method comprises a multi-mode item set construction module, a similarity measurement module and a recognition result decision module;
the multi-mode item set construction module is used for respectively constructing historical data and user data to be identified into an expandable multi-mode item set;
the similarity measurement module is used for carrying out similarity measurement on the extensible multi-mode item set of the user data to be identified and the extensible multi-mode item set of the pre-constructed historical data;
the recognition result decision module is used for making a decision on the user recognition result according to the similarity measurement result to obtain the recognition result.
Preferably, the recognition result decision module is further configured to obtain a recognition result according to the similarity measurement result by using a decision method based on a similarity measurement, or obtain a user recognition result by using an FI method.
Compared with the prior art, the application has the following beneficial effects:
the application constructs the expandable multi-mode item set according to the item sequence (user big data) of the user, thereby achieving the purpose of user identification without introducing other additional user information. The similarity measurement method provided by the application can be used for measuring the similarity more effectively by combining the two general similarity distances. The decision scheme based on various similarity measures further improves the accuracy of user identification at the cost of a certain rejection rate. The method has important significance in information evidence collection and personalized recommendation, and also provides a new challenge for privacy protection of network space.
Drawings
FIG. 1 is a flow chart of a method of user identification based on a multimodal set of user data.
FIG. 2 is a block diagram of a user identification system based on a user behavior big data extensible multi-modal set.
Detailed Description
In order to more clearly illustrate the purposes and technical schemes of the application, a user identity recognition method and a system based on a multi-mode item set of user data are further described below by combining an embodiment and a drawing. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not limiting.
As shown in fig. 1, the present application provides a user identity recognition method based on a multi-mode item set of user data, and the embodiment is as follows:
in this embodiment, the history data and the user data to be identified are online shopping data sets, each record is record= < userId, date, time, and product id >, where userId is a user number, date and time are recorded date and specific time, respectively, and product id represents the number of the commodity purchased by the user. The product IDs of each user are sequentially fetched in time sequence to form a project sequence.
The item sequence of the ith user of the history data x is noted as: for user u xi The number of records; the sequence of items of the j-th user of the user data y to be identified is noted as: for user u yj Is a record of the number of bars. Taking the modulo n=3 of the extensible item set, there is:
taking r top ＝0.5，
According to the formulaAnd
processing the multimode item to obtain
The similarity measure JKL is used below, namely:calculation u xi And u yj The formula +.>Or formula (la)
And making a decision on the user identification result according to the obtained similarity. For the case of using a similarity measurement method, for example, one of Jaccard coefficients, KL divergence, cosine similarity, pearson correlation coefficients, and the like is selected, the formula is requiredAnd obtaining a recognition result.
For the case of using multiple similarity metric methods, the FI decision scheme is used to derive the recognition result. In this embodiment, the modulo n=3 of the extensible item set may be based on two types of sub-schemes of the FI decision scheme, i.e., formulasOr formula (la)And obtaining a recognition result.
Specifically, two cases are divided:
(1) The similarity measurement method of the homogeneous mode comprises the following steps: or->At this time, k=2, indicating that two similarity measurement methods (Jaccard coefficient and KL divergence) are used; n is 1,2 and 3 respectively.
(2) A multi-class modulus similarity measurement method, namely:at this time, k=2, which means that two similarity measurement methods (Jaccard coefficient and KL divergence) are used, n=2, which means that N is 1 and 2, respectively; or alternativelyAt this time, k=2 indicates that two similarity measurement methods (Jaccard coefficient and KL divergence) are used, and n=3 indicates that N is 1,2, and 3, respectively.
Can calculate the formula according to the precisionFormula of rejection rateUser recognition effects are evaluated using a variety of similarity measures.
As shown in fig. 2, the application also provides a user identity recognition system of the multimode item set based on the user data, which comprises a multimode item set construction module, a similarity measurement module and a recognition result decision module;
the multi-mode item set construction module is used for respectively constructing an extensible multi-mode item set by historical data and user data to be identified;
the similarity measurement module calculates similarity according to the historical data and the extensible multi-mode item set of the user data to be identified;
the recognition result decision module obtains a recognition result by using a decision method based on a similarity measure according to the similarity measure result, or obtains a user recognition result by using an FI method.
The above examples are preferred embodiments of the present application, and the embodiments of the present application are not limited thereto. Any modification, substitution, etc. of the present application are included in the scope of protection of the present application as long as they do not depart from the spirit and scope of the present application.

Claims (6)

1. A user identification method based on a multi-mode item set of user data, comprising the steps of:
constructing an expandable multi-mode item set of user data to be identified;
performing similarity measurement on the extensible multi-mode item set of the user data to be identified and the extensible multi-mode item set of the pre-constructed historical data;
making a decision on the user identification result according to the similarity measurement result to obtain the identification result;
the method for measuring the similarity between the extensible multi-mode item set of the user data to be identified and the extensible multi-mode item set of the pre-constructed historical data specifically comprises the following steps:
the ith user in the history data is noted as u xi Its corresponding extensible multi-modal set is denoted asThe j-th user in the user data to be identified is marked as u yj Its corresponding extensible multimode item set is denoted +.>The similarity measure is defined as:
this method is called JKL, wherein,represents u calculated by JKL xi And u yj Similarity of (2); />Representing u calculated using Jaccard coefficients xi And u yj The similarity of (2), the formula is:
wherein,representation->And->Length of intersection, i.e.)> Representation->And->Length of union, i.e.)>
Represents u calculated by KL method xi And u yj The similarity of (2), the formula is:
wherein,and->Respectively indicate->And->Frequency histogram of intersection, D (||·) is KL divergence function expressed as:
wherein k= |α= |β| represents the length of α or β;
the user identification result in making a decision on the user identification result according to the similarity measurement result is defined as:
wherein V is y ＝{u yj I j e {1,2,.,... ij Is u xi And u yj Similarity of (i.e. w) ij ＝similarity(u xi ，u yj )；
The decision making of the user identification result comprises two decision schemes:
I. decision scheme based on a similarity measure: the identification result can be obtained by applying a similarity measurement method, and the formula is as follows:
wherein k represents a certain similarity measurement method, and Jaccard coefficients, KL divergence, cosine similarity and Pearson correlation coefficients can be selected;
decision scheme based on multiple similarity metrics: the scheme is called FI, is a fusion decision method for intersecting candidate sets generated based on various similarity measurement methods, and is defined as the formula:
where k.epsilon. {1, 2.. The number of similarity measures used is denoted by K, and intersecting the results obtained by the K similarity measurement methods to obtain a final recognition result.
2. The method for identifying a user identity based on a multimodal set of user data according to claim 1, wherein the steps of constructing an extensible multimodal set of user data to be identified and constructing an extensible multimodal set of history data are both MISFUB, the MISFUB specifically comprising:
s2.1, constructing an extensible multimode item; the behavior, state or event of the user is recorded as Item, and the Item sequence of the user is recorded as ItemSequence＝{Item 1 ，Item 2 ，Item 3 ，......，Item Nr -where Nr is a positive integer;
the modulus of the extensible item set is denoted as n; wherein the level 1 Item set is the Item Sequence in the natural state and is marked as C 1 ＝{S i I=1, 2, &..; level 2 item set is denoted as C 2 ＝{S ij I=1, 2,) Nr-1 and j=i+1, & gt, nr; level 3 item set is denoted as C 3 ＝{S ijk I=1, 2,) Nr-2 and j=i+1,) Nr-1 and k=i+2, & Nr; n-level item set is denoted as C n ＝{S i1i2...in |i 1 =1, 2.,. Nr-n+1 and i k ＝i l +k-1，k＝1，2，......，n}；S i1i2...in Representing a sequence of n items from Nr items;
s2.2, summing the n multimode item sets obtained in the step S2.1, namely:
s2.3 pairIs ordered in descending order according to the frequency of occurrence of each sequence S, namely:wherein->Is->The frequency of the mth sequence of (c),
s2.4, define the ratio r top The method comprises the following steps:
according to r top Extraction ofThe first sequences of the middle row form F n The method comprises the following steps:
3. the method for user identification based on a multimodal set of user data according to claim 1, wherein the decision scheme based on a plurality of similarity metrics comprises two types of fused decision schemes:
(1) The similarity of the same model fuses the decision, the formula is defined as:
wherein K is {1,2,.. The number of the similarity measurement method used is denoted by K, n is a modulus of the expandable item set, and a positive integer is taken;
(2) A multi-class modulus similarity fusion decision, the formula is defined as:
where k.epsilon. {1, 2.. The number of similarity measures used is denoted by K, N e {1, 2.,. N represents the modulo taking of the extensible item set for N different cases.
4. A method for identifying a user identity based on a multimodal set of user data as claimed in claim 3, wherein the accuracy calculation formula of the decision scheme based on a plurality of similarity measures is defined as:
wherein V is x ＝{u xi I e {1,2,.,. Nu }, represents a set of Nu users in the history data;
because ofPossibly empty, the reject rate is defined as:
5. the user identity recognition system based on the multimode item set of the user data is characterized by comprising a multimode item set construction module, a similarity measurement module and a recognition result decision module;
the multi-mode item set construction module is used for respectively constructing historical data and user data to be identified into an expandable multi-mode item set;
the similarity measurement module is used for carrying out similarity measurement on the extensible multi-mode item set of the user data to be identified and the extensible multi-mode item set of the pre-constructed historical data;
the recognition result decision module is used for making a decision on the user recognition result according to the similarity measurement result to obtain a recognition result;
the method for measuring the similarity between the extensible multi-mode item set of the user data to be identified and the extensible multi-mode item set of the pre-constructed historical data specifically comprises the following steps:
the ith user in the history data is noted as u xi Its corresponding extensible multi-modal set is denoted asThe j-th user in the user data to be identified is marked as u yj Its corresponding extensible multimode item set is denoted +.>The similarity measure is defined as:
this method is called JKL, wherein,represents u calculated by JKL xi And u yj Similarity of (2); />Representing u calculated using Jaccard coefficients xi And u yj The similarity of (2), the formula is:
wherein,representation->And->Length of intersection, i.e.)>Representation->And->Length of union, i.e.)>
Represents u calculated by KL method xi And u yj The similarity of (2), the formula is:
wherein,and->Respectively indicate->And->Frequency histogram of intersection, D (||·) is KL divergence function expressed as:
wherein k= |α= |β| represents the length of α or β;
the user identification result in making a decision on the user identification result according to the similarity measurement result is defined as:
wherein V is y ＝{u yj I j e {1, 2.,. Nu } "represents the set of Nu users, w, in the user data to be identified ij Is u xi And u yj Similarity of (i.e. w) ij ＝similarity(u xi ，u yj )；
The decision making of the user identification result comprises two decision schemes:
I. decision scheme based on a similarity measure: the identification result can be obtained by applying a similarity measurement method, and the formula is as follows:
wherein k represents a certain similarity measurement method, and Jaccard coefficients, KL divergence, cosine similarity and Pearson correlation coefficients can be selected;
decision scheme based on multiple similarity metrics: the scheme is called FI, is a fusion decision method for intersecting candidate sets generated based on various similarity measurement methods, and is defined as the formula:
wherein, K epsilon {1,2, & gt..A.K.. Represents the number of the similarity measurement method used, and the results obtained by the K similarity measurement methods are intersected to obtain the final recognition result.
6. The system of claim 5, wherein the decision module is further configured to derive the recognition result using a similarity metric-based decision method or derive the user recognition result using an FI method based on a similarity metric based result.
CN202011109382.3A 2020-10-16 2020-10-16 User identity recognition method and system based on multi-mode item set of user data Active CN112381112B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011109382.3A CN112381112B (en) 2020-10-16 2020-10-16 User identity recognition method and system based on multi-mode item set of user data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011109382.3A CN112381112B (en) 2020-10-16 2020-10-16 User identity recognition method and system based on multi-mode item set of user data

Publications (2)

Publication Number Publication Date
CN112381112A CN112381112A (en) 2021-02-19
CN112381112B true CN112381112B (en) 2023-11-07

Family

ID=74579895

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011109382.3A Active CN112381112B (en) 2020-10-16 2020-10-16 User identity recognition method and system based on multi-mode item set of user data

Country Status (1)

Country Link
CN (1) CN112381112B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113722601B (en) * 2021-09-07 2024-04-16 南方电网数字电网研究院股份有限公司 Power measurement information recommendation method, device, computer equipment and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107423754A (en) * 2017-06-28 2017-12-01 山东航天电子技术研究所 A kind of radiation source automatic recognition system based on the more attribute autonomous intelligence decision-makings of parameter
CN109684538A (en) * 2018-12-03 2019-04-26 重庆邮电大学 A kind of recommended method and recommender system based on individual subscriber feature
CN109840702A (en) * 2019-01-24 2019-06-04 山西开拓科技股份有限公司 A kind of new projects' collaborative recommendation method based on multi-core integration
CN110096499A (en) * 2019-04-10 2019-08-06 华南理工大学 A kind of the user object recognition methods and system of Behavior-based control time series big data
CN111242218A (en) * 2020-01-13 2020-06-05 河南科技大学 Cross-social network user identity recognition method fusing user multi-attribute information

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107423754A (en) * 2017-06-28 2017-12-01 山东航天电子技术研究所 A kind of radiation source automatic recognition system based on the more attribute autonomous intelligence decision-makings of parameter
CN109684538A (en) * 2018-12-03 2019-04-26 重庆邮电大学 A kind of recommended method and recommender system based on individual subscriber feature
CN109840702A (en) * 2019-01-24 2019-06-04 山西开拓科技股份有限公司 A kind of new projects' collaborative recommendation method based on multi-core integration
CN110096499A (en) * 2019-04-10 2019-08-06 华南理工大学 A kind of the user object recognition methods and system of Behavior-based control time series big data
CN111242218A (en) * 2020-01-13 2020-06-05 河南科技大学 Cross-social network user identity recognition method fusing user multi-attribute information

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于分层识别的快速说话人识别研究;茅正冲;涂文辉;;计算机工程与科学(第07期);第1-4页 *

Also Published As

Publication number Publication date
CN112381112A (en) 2021-02-19

Similar Documents

Publication Publication Date Title
CN104615775B (en) A kind of user&#39;s recommendation method and device
Xu et al. Integrated collaborative filtering recommendation in social cyber-physical systems
US20130132481A1 (en) Method, apparatus and system for providing social network service using social activities
KR102106203B1 (en) Expert recommending method and system for providing social network system based question and answer service
JPWO2012118087A1 (en) Recommender system, recommendation method, and program
Kommineni et al. Machine learning based efficient recommendation system for book selection using user based collaborative filtering algorithm
CN106815347A (en) Improvement slope one Collaborative Filtering Recommendation Algorithms based on commodity similarity
US20160004937A1 (en) System and method for determining string similarity
CN111782927B (en) Article recommendation method and device and computer storage medium
CN115082041B (en) User information management method, device, equipment and storage medium
JP2014203442A (en) Recommendation information generation device and recommendation information generation method
CN104933595A (en) Collaborative filtering recommendation method based on Markov prediction model
CN111400612A (en) Personalized recommendation method integrating social influence and project association
CN112381112B (en) User identity recognition method and system based on multi-mode item set of user data
KR101458588B1 (en) Expert curation recommendation system and expert recommendation method using thereof by field
CN111428145A (en) Recommendation method and system fusing tag data and naive Bayesian classification
JP5331159B2 (en) Product recommendation method, product recommendation device, and product recommendation program
Saraswathi et al. Deep Learning Enabled Social Media Recommendation Based on User Comments.
CN116861076A (en) Sequence recommendation method and device based on user popularity preference
CN104537080B (en) Information recommends method and system
CN113886697A (en) Clustering algorithm based activity recommendation method, device, equipment and storage medium
Eyal et al. Identifying missing node information in social networks
Kumar et al. Point-of-interest based classification of similar users by using support vector machine and status homophily
Singh et al. Improving customer segmentation in e-commerce using predictive neural network
Qiu A predictive model for customer purchase behavior in e-commerce context

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant