Network access behavior characteristic group dynamic mining method and system under enhanced condition

Abstract

The invention belongs to the field of data mining, and specifically relates to a network access behavior characteristic group dynamic mining method and system under an enhancement condition, being characterized in that an intelligent model capable of quickly and efficiently searching maximum dichotomous groups in dynamically changed data is established aiming at statistical data of individual access web pages under the continuous change condition that the relationship between individuals and access is increased. The scheme is as follows: an input interface is provided for a user; the user inputs effective frequency statistical data of accessing each webpage by an individual to convert the data into a matrix form, scans once on the basis of the matrix to search all maximum dichotomies in the matrix and stores the maximum dichotomies in a memory; an interface for inputting the matrix point or edge incremental data is provided for the user, so as to normalize the incremental data input bythe user into the incremental data of the edge; and finally, an iterative search process is executed on each piece of data, and all maximum dichotomials obtained by the last iteration are output, namely all groups with maximized common network access behavior characteristics.

Description

technical field [0001] The invention belongs to the field of data mining, and in particular relates to a method and system for dynamically mining network access behavior characteristic groups under enhanced conditions. Background technique [0002] At present, relationship graphs are widely used in scientific fields such as social networks, gene biology, and cognitive radio. In many big data domains, there is a need to search for groups or objects that maximize common characteristics. Groups or targets and their characteristics are usually expressed abstractly in the form of various graphs. Among them, groups or targets with the maximum common characteristics are expressed in the form of a special graph, including: maximum clique, maximum dichotomous clique, quasi dichotomous clique, Maximum edge dichotomy, maximum balanced dichotomy, frequent itemsets, etc. [0003] The present invention mainly aims at the online network access relationship, from which to search for group...

AI Technical Summary

Problems solved by technology

Although these methods can search for the largest binary group in dynamically changing data, this method based on sliding windows is inherently limited by the size of the window, so the results obtained are often rough rather than precise.

Images