A recommendation system and method for measuring node intimacy in a directed weighted graph

A technology of node intimacy and recommendation system, applied in the field of data mining, it can solve the problems of reducing user query efficiency and reducing the accuracy of final recommendation results, achieving the effects of wide applicability, simple implementation and low computational complexity

Inactive Publication Date: 2016-08-31
北京航空航天大学深圳研究院
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These methods often only consider the weight of the edges in the graph and ignore the importance of the nodes themselves in the calculation process, which reduces the accuracy of the final recommendation results and reduces the efficiency of user queries.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A recommendation system and method for measuring node intimacy in a directed weighted graph
  • A recommendation system and method for measuring node intimacy in a directed weighted graph
  • A recommendation system and method for measuring node intimacy in a directed weighted graph

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0064] The node intimacy measure in the weighted graph can be used in many different fields. For example, in a social network, link prediction can be performed through the intimacy between users, that is, to predict the future relationship between two unfamiliar users based on existing data. Whether a connection will occur at some point.

[0065] In a social network, users interact through sharing, commenting, and "@" operations, and the scores for each operation are s(r)=3, s(c)=1, and s(α)=5. Therefore, in a social network, the weight of an edge represents the frequency of interaction between two users. If the number of times user A shares, comments, and "@" operations on user B is s, c, and a respectively, then the weight of the edge The calculation formula is:

[0066] I(A→B)=s·s(s)+c·s(c)+a·s(a)

[0067] In a social network, the importance of a user is composed of two aspects: user activity and user influence, where user activity is obtained by the number of times the u...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A recommendation system and method for measuring node intimacy in a directed weighted graph, (1) Obtaining the weight of an edge in a given graph: According to the user's needs, give the definition of the weight of an edge in the graph, and calculate the edge according to this definition (2) Normalize the weights of the edges in the given graph: After obtaining the weights of the edges, normalize the weights of the edges so that the intimacy between non-directly connected nodes can be calculated later; (3) Obtain the importance measure of the nodes in the given graph: Calculate the importance of the nodes in the graph. Under the same topology, the intimacy between the important nodes and the given nodes should also be high; (4) in Add new virtual nodes to a given graph: add edges between virtual nodes and existing nodes, and assign weights to edges according to node importance (5) Calculate the final result of intimacy between nodes: on the modified weighted graph, Using the restart random walk algorithm, the intimacy between the specified node and other nodes is calculated.

Description

technical field [0001] The invention relates to a recommendation system and method for measuring node intimacy in a directed weighted graph, belonging to the technical field of data mining. Background technique [0002] In scientific research, problems in many fields, including SNS and biomedicine, can be abstracted as graph theory problems. At present, many applications are dealing with this kind of graph-based knowledge representation. In these applications, k The relationship between entities of interest is often of great significance. For example, in the field of public security, by analyzing the relationship between multiple suspects, it is possible to find the core criminal who has a strong connection with all suspects. In the field of biomedicine , by analyzing the relationship between genes and proteins, we can discover the effect of a certain drug on the disease, so as to get a new treatment method for the disease. [0003] Computing the affinity between nodes in a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F19/00G06F17/30
Inventor李超吕志强赵彩贝朱耿良
Owner北京航空航天大学深圳研究院