Method for analyzing user behaviors in wireless local area network

Abstract

The invention discloses a method for analyzing user behaviors in a wireless local area network. The method comprises the following steps of: acquiring the network access information of a user in the wireless local area network, dividing the network access information of the user according to a time interval, so as to generate divided network access records, sampling the divided network access records, so as to generate sampled network access records, calculating dissimilarities between each two of all the users in the wireless local area network, and clustering the dissimilarities among all the users in the wireless local area network, so as to obtain a clustering result of the user behaviors. By using the method, the user behaviors in a wireless network represent the sequence property and the distance property of the movement of the user, so as to improve the sensitiveness of a distance and the sensitiveness of time, and meanwhile, the method is superior to the prior art in time complexity.

Description

technical field [0001] The invention belongs to the field of wireless local area network, and more specifically relates to a method for analyzing user behavior in the wireless local area network. Background technique [0002] In order to measure the distance of user behavior in the wireless network more accurately, it is necessary to mine the behavior between users through network data. Wei-jen Hsu and Ahmed Helmy proposed a method of mining behavior between users, which is based on the measurement of the average minimum vector distance (AMVD) algorithm, and uses the user's behavior matrix to represent the user, and through This matrix calculates user dissimilarity, thereby accurately defining user behavior characteristics. [0003] However, there are four problems in the existing user behavior analysis methods. First, it uses a matrix to represent data, which erases the order of user movement and will lead to deviations in the actual meaning of the clustering results; seco...

AI Technical Summary

Problems solved by technology

[0003] However, there are four problems in the existing user behavior analysis methods. First, it uses a matrix to represent data, which erases the order of user movement and will lead to deviations in the actual meaning of the clustering results; second, normalization Finally, the proportion matrix of online time ignores the length of users’ online time; the third

The measurement based on the AMVD algorithm is too sensitive; fourth, the AMVD algorithm does not conform to the triangle inequality, which makes the analysis method unsuitable for many data mining and machine learning algorithms

Images