CN108063810A - A kind of recommendation method based on the filtering of network part structural information - Google Patents

A kind of recommendation method based on the filtering of network part structural information Download PDF

Info

Publication number
CN108063810A
CN108063810A CN201711318237.4A CN201711318237A CN108063810A CN 108063810 A CN108063810 A CN 108063810A CN 201711318237 A CN201711318237 A CN 201711318237A CN 108063810 A CN108063810 A CN 108063810A
Authority
CN
China
Prior art keywords
node
mrow
nodes
network
similarity
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
Application number
CN201711318237.4A
Other languages
Chinese (zh)
Inventor
杨旭华
徐恩平
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.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
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 Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN201711318237.4A priority Critical patent/CN108063810A/en
Publication of CN108063810A publication Critical patent/CN108063810A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/147—Network analysis or design for predicting network behaviour
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/02—Network architectures or network communication protocols for network security for separating internal from external traffic, e.g. firewalls
    • H04L63/0227—Filtering policies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/535—Tracking the activity of the user

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Security & Cryptography (AREA)
  • Computing Systems (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

A kind of recommendation method based on network part information filtering, comprises the following steps：Live network structural information is obtained, establishes network model G (V, E)；Determine destination node vi, use θiRepresent all possible connecting node set to be recommended；In set θiIn appoint take a node vj, obtain node viAnd vjCommon neighbor node set Ψ (i, j)；Choose node v successively in set Ψ (i, j)h, calculate node vhDegree kh, node viAnd vhCommon neighbours' number, node vjAnd vhCommon neighbours' number；Calculate node viAnd vjIndex of similarity；Set of computations θiIn all node and viIndex of similarity, it is highest three nodes of connection possibility that access, which is worth highest three nodes,.The present invention can effectively improve link prediction arithmetic accuracy.

Description

A kind of recommendation method based on the filtering of network part structural information
Technical field
The present invention relates to network recommendation technical fields, particularly relate to a kind of recommendation based on the filtering of network part structural information Method.
Background technology
Computer, internet and the web technology of rapid development change people’s lives, and people make friends in virtual community Good friend, in news website browse news, in video website watch film, in virtual library consult books, in electric business Article is bought in platform.But people have also experienced the worry that information expansion is brought, i.e. people while rich Color Life is enjoyed Maximally related information can not be fast and effeciently found in mass data.The data volume of the information such as film, books, webpage is easily With millions, the growth rate of these data messages is considerably beyond the natural processing capacity of the mankind.In this big data Background under, user obtain information needed cost it is increasing, rely solely on the mode of traditional manpower can not evaluate and Select these articles.
In this case, the most attractive method of effectively filtering magnanimity information is exactly commending system.It is basis The interested information of user, product etc. are recommended the recommendation of personalized information system of user by information requirement, interest of user etc. System.Commending system carries out personalized calculating by studying the interest preference of user, by the point of interest of system discovery user, so as to Guiding user has found the information requirement of oneself.One good commending system can not only provide personalized service to the user, moreover it is possible to Substantial connection is established between user, allows user to recommending to generate dependence.Researchers propose various proposed algorithm, In, the proposed algorithm based on link prediction has received widespread attention.
Link prediction in network refers to, how by information such as known network structures, predict that not yet generation connects in network The possibility of connection is generated between two nodes connect.Vertex in network represents user, while customer relationship is represented, link prediction Problem is exactly the analysis to user's future relation.
The content of the invention
In order to overcome the shortcomings of that the precision of prediction of existing recommendation method is relatively low, the present invention is by believing network partial structurtes The research of breath, it is proposed that a kind of link prediction algorithm based on network part information filtering, and the algorithm is applied to recommendation system In system, it is proposed that a kind of recommendation method based on network part information filtering for effectively improving link prediction arithmetic accuracy.
Technology comprises the concrete steps that used by the present invention solves its technical problem：
A kind of recommendation method based on the filtering of network part structural information, comprises the following steps：
Step 1：Live network structural information is obtained, establishes network model G (V, E), V represents the node in network, E generations Company side in table network；
Step 2：Determine destination node vi, it is node viRecommend the highest node of connection possibility, use θiRepresent it is all can The connecting node set to be recommended of energy；
Step 3：In set θiIn appoint take a node vj, obtain node viAnd vjCommon neighbor node set Ψ (i, J)=Γ (i) ∩ Γ (j), wherein Γ (i) and Γ (j) represent node v respectivelyiAnd vjNeighbor node set；
Step 4：Node v is arbitrarily chosen in set Ψ (i, j)h, calculate node vhDegree kh, node viAnd vhIt is common Neighbours' numberNode vjAnd vhCommon neighbours' number
Step 5：Calculate node viAnd vjIndex of similarity
Step 6：To set θiIn all node, repeat step 3 to step 5, calculate all nodes and viIt is similar Index to be spent, all nodes are arranged from big to small according to corresponding index of similarity numerical value, access is worth highest three nodes, this Three nodes are highest three nodes of connection possibility.
Beneficial effects of the present invention are：From actual life the phenomenon that friend recommendation, in classical common neighbours side On the basis of method, unrelated local network structure information is filtered out, is effectively improved the precision of link prediction algorithm.
Description of the drawings
Fig. 1 is the partial structural diagram of a network model.
Specific embodiment
The present invention will be further described below in conjunction with the accompanying drawings.
With reference to Fig. 1, a kind of recommendation method based on the filtering of network part structural information comprises the following steps：
Step 1：Live network structural information is obtained, establishes network model G (V, E), V represents the node in network, E generations Company side in table network, such as the partial structural diagram that Fig. 1 is a network model；
Step 2：Determine destination node vi, it is node v such as the node in Fig. 1 1.iRecommend the highest section of connection possibility Point, uses θiRepresent all possible connecting node set to be recommended, such as the node in Fig. 1 2., node 5., node 6. belong to section The connecting node set θ to be recommended of point 1.i；
Step 3：In set θiIn appoint take a node vj, obtain node v such as the node in Fig. 1 2.iAnd vjCommon neighbour It occupies node set Ψ (i, j)=Γ (i) ∩ Γ (j), wherein Γ (i) and Γ (j) and represents node v respectivelyiAnd vjNeighbor node collection Close, as common neighbor node of the node 1. with node 2. have node 3. with node 4.；
Step 4：Node v is arbitrarily chosen in set Ψ (i, j)h, calculate node vhDegree kh, node viAnd vhIt is common Neighbours' numberNode vjAnd vhCommon neighbours' numberSuch as the k of node 3.h=5,
Step 5：Calculate node viAnd vjIndex of similarity
Practical significance representated by the node index of similarity of this method is two nodes to be calculated with common neighbor node The probability connected for medium.Specifically, " 2 (k in denominatorh- 1) institute -1 " is represented it is possible that connecting number,WithIt represents Already present connection number.
As index of similarity of the node 1. with node 2. is：
Step 6：To set θiIn all node, repeat step 3 to step 5, the similarity for calculating all nodes refers to Mark, all nodes are arranged from big to small according to its index of similarity numerical value, and access is worth highest three nodes, these three nodes As highest three nodes of connection possibility.

Claims (1)

  1. A kind of 1. recommendation method based on the filtering of network part structural information, it is characterised in that：Comprise the following steps：
    Step 1：Live network structural information is obtained, establishes network model G (V, E), V represents the node in network, and E represents net Company side in network；
    Step 2：Determine destination node vi, it is node viRecommend the highest node of connection possibility, use θiRepresent all possible Connecting node set to be recommended；
    Step 3：In set θiIn appoint take a node vj, obtain node viAnd vjCommon neighbor node set Ψ (i, j)=Γ (i) ∩ Γ (j), wherein Γ (i) and Γ (j) represent node v respectivelyiAnd vjNeighbor node set；
    Step 4：Node v is arbitrarily chosen in set Ψ (i, j)h, calculate node vhDegree kh, node viAnd vhCommon neighbours NumberNode vjAnd vhCommon neighbours' number
    Step 5：Calculate node viAnd vjIndex of similarity
    <mrow> <mi>S</mi> <mi>i</mi> <mi>m</mi> <mo>=</mo> <munder> <mo>&amp;Sigma;</mo> <mrow> <msub> <mi>v</mi> <mi>h</mi> </msub> <mo>&amp;Element;</mo> <mi>&amp;Gamma;</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>&amp;cap;</mo> <mi>&amp;Gamma;</mi> <mrow> <mo>(</mo> <mi>j</mi> <mo>)</mo> </mrow> </mrow> </munder> <mfrac> <mn>1</mn> <mrow> <mn>2</mn> <mrow> <mo>(</mo> <msub> <mi>k</mi> <mi>h</mi> </msub> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>-</mo> <mn>1</mn> <mo>-</mo> <msubsup> <mi>S</mi> <mrow> <mi>i</mi> <mi>h</mi> </mrow> <mrow> <mi>C</mi> <mi>N</mi> </mrow> </msubsup> <mo>-</mo> <msubsup> <mi>S</mi> <mrow> <mi>j</mi> <mi>h</mi> </mrow> <mrow> <mi>C</mi> <mi>N</mi> </mrow> </msubsup> </mrow> </mfrac> </mrow>
    Step 6：To set θiIn all node, repeat step 3 to step 5, calculate all nodes and viSimilarity refer to Mark, all nodes are arranged from big to small according to corresponding index of similarity numerical value, and access is worth highest three nodes, these three Node is highest three nodes of connection possibility.
CN201711318237.4A 2017-12-12 2017-12-12 A kind of recommendation method based on the filtering of network part structural information Pending CN108063810A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711318237.4A CN108063810A (en) 2017-12-12 2017-12-12 A kind of recommendation method based on the filtering of network part structural information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711318237.4A CN108063810A (en) 2017-12-12 2017-12-12 A kind of recommendation method based on the filtering of network part structural information

Publications (1)

Publication Number Publication Date
CN108063810A true CN108063810A (en) 2018-05-22

Family

ID=62138300

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711318237.4A Pending CN108063810A (en) 2017-12-12 2017-12-12 A kind of recommendation method based on the filtering of network part structural information

Country Status (1)

Country Link
CN (1) CN108063810A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020000724A1 (en) * 2018-06-29 2020-01-02 平安科技（深圳）有限公司 Method, electronic device and medium for processing communication load between hosts of cloud platform
WO2022017082A1 (en) * 2020-07-24 2022-01-27 北京沃东天骏信息技术有限公司 Method and apparatus for detecting false transaction orders

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160292303A1 (en) * 2015-04-03 2016-10-06 Oracle International Corporation Distributed graph processing system that support remote data read with proactive bulk data transfer
CN106817251A (en) * 2016-12-23 2017-06-09 烟台中科网络技术研究所 A kind of link prediction method and device based on node similarity

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160292303A1 (en) * 2015-04-03 2016-10-06 Oracle International Corporation Distributed graph processing system that support remote data read with proactive bulk data transfer
CN106817251A (en) * 2016-12-23 2017-06-09 烟台中科网络技术研究所 A kind of link prediction method and device based on node similarity

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ZHAO YANG: "An Improved Link Prediction Algorithm Based on", 《2016 7TH IEEE INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING AND SERVICE SCIENCE (ICSESS)》 *
尹永超: "链路预测中一种改进的相似度指标算法", 《小型微型计算机系统》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020000724A1 (en) * 2018-06-29 2020-01-02 平安科技（深圳）有限公司 Method, electronic device and medium for processing communication load between hosts of cloud platform
WO2022017082A1 (en) * 2020-07-24 2022-01-27 北京沃东天骏信息技术有限公司 Method and apparatus for detecting false transaction orders

Similar Documents

Publication Publication Date Title
Depren et al. Energy consumption and environmental degradation nexus: A systematic review and meta-analysis of fossil fuel and renewable energy consumption
CN102789462B (en) A kind of item recommendation method and system
Weber et al. The flow of digital news in a network of sources, authorities, and hubs
CN103092911B (en) A kind of mosaic society label similarity is based on the Collaborative Filtering Recommendation System of k nearest neighbor
CN103455515B (en) User recommendation method and system in SNS (social networking services) community
CN112131480B (en) Personalized commodity recommendation method and system based on multilayer heterogeneous attribute network representation learning
CN103559197A (en) Real-time music recommendation method based on context pre-filtering
CN101271559A (en) Cooperation recommending system based on user partial interest digging
CN103729359A (en) Method and system for recommending search terms
CN103258020A (en) Recommending system and method combining SNS and search engine technology
CN103761237A (en) Collaborative filtering recommending method based on characteristics and credibility of users
CN104462383A (en) Movie recommendation method based on feedback of users&#39; various behaviors
CN103309972A (en) Recommend method and system based on link prediction
CN103020712B (en) A kind of distributed sorter of massive micro-blog data and method
CN105023178B (en) A kind of electronic commerce recommending method based on ontology
Dev et al. Recommendation system for big data applications based on set similarity of user preferences
CN107145541B (en) Social network recommendation model construction method based on hypergraph structure
CN110083764A (en) A kind of collaborative filtering cold start-up way to solve the problem
CN110335123A (en) Method of Commodity Recommendation, system, computer-readable medium and device based on social electric business platform
CN112102029A (en) Knowledge graph-based long-tail recommendation calculation method
CN116166875A (en) Bidirectional cross-domain recommendation method of heterogeneous graph neural network based on element path enhancement
Honarparvar et al. Improvement of a location-aware recommender system using volunteered geographic information
CN108063810A (en) A kind of recommendation method based on the filtering of network part structural information
CN104462597B (en) A kind of positive negativity of synthetic user scores and the collaborative filtering method of scoring preference heterogeneity
CN107920133A (en) A kind of recommendation method based on ternary closure information in localized network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20180522