Geographical social network based user similarity computation method

Abstract

The invention relates to analysis of social network user recommendation and information service recommendation technologies, and more particularly to a geographical social network based user similarity computation method, belonging to the field of public opinion monitoring. The method mainly comprises following steps of space time semantic meaning extracting, user space time access model establishing and user similarity computing. The method has the beneficial effects that space time semantic meaning division is more comprehensive and the problem caused by data sparsity is solved; due to consideration of impact of both location function semantic meaning and geographical factor on user similarity, user portrait modeling is more comprehensive; and social network user similarity computation is carried out by combination of physical space time vicinity of traces and semantic meaning space time similarity, so that accuracy is higher, the social network user groups are divided, and the use classification and interest type determination effect are better.

Description

technical field [0001] The invention belongs to the field of public opinion monitoring, relates to social network user recommendation and commercial service recommendation technical analysis, and in particular to a method for calculating user similarity under geographic social networks. Background technique [0002] With the popularization of intelligent mobile terminals with mobile positioning functions and the development of online social networks, location services and online social networks are tending to merge, resulting in LBSN (Location-basedonlineSocialNetwork). Because users can record their geographic behaviors and their feelings about geographic events and social functions in real time through LBSN, LBSN not only reflects the virtual relationship and connection between residents, but also reflects the activities of urban entities in cyberspace. An important research direction of current LBSN mining is user similarity calculation. Due to differences in income leve...

AI Technical Summary

Problems solved by technology

[0007] (2) Lack of a more comprehensive semantic time division method

It does not consider the overall laws of social work and rest, and cannot recognize time that is not the same physical moment but has the same semantic meaning, such as working hours, holidays, etc.; some documents artificially set the interval of each semantic period, with greater randomness , these time slots cannot reflect the social and cultural meaning of time, nor can they reflect the activity differences among users to the greatest extent.

[0008] (3) There is a lack of methods to properly express the intensity of location visits in different time periods

It does not take into account the difference in the activity level and number of check-ins of users at different times, and ignores the difference between users who have similar location arrival patterns over a long period of time but have prominent differences in location arrival time

[0009] (4) Lack of sequential pattern of location access that expresses statistical significance of users over a long period of time

Another type of research is to use all the user's location access data as an object, and use the LDA model to compare the object similarity, which can obtain the location access intensity in a statistical sense in the global time, but does not consider the user's location in each time period. Statistical strength of position occurrence

Images