Social Relationship Mining And Verificating Framework Based On Short-range Wireless Data

As the Internet technology develops,people using the network to communicate has become a mainstream communication method.This social network formed through social networking has attracted widespread attention from researchers.The research found that some people's behavioral habits and friends' hobbies in real life will also have impact on their social interaction on the Internet.This shows that there is a lot of social information hidden in the social network formed by the network.How to dig out these hidden social information is also one of the key research contents in this paper.Due to the different types of networks,the types of social networks are also diverse.Online social networks,traditional social networks,and social networks in short-range wireless networks are mainly involved in this paper.At present,a single study on online social networks or short-range wireless networks has become a research hotspot.But comprehensive research on these two types of networks is still relatively scarce.Whether there is a correlation between these two networks has not reached a unified conclusion in the scientific community.So in this article we will not discuss too much about the existence of association between these two network types,but propose a specific framework based on short-range wireless data,which can mine some social relationships hidden in short-range wireless networks.And map the mined relationships to online social networks or traditional social networks.According to the extracted number of data indicators and the relationship between nodes,we propose mapping based on single index,mapping based on multiple metrics,direct relationship mapping and indirect relationship mapping.And combining the mapping based on the threshold,mapping based on the accumulation values,mapping based on the accumulation values and mapping based on auxiliary data designed according to different data characteristics,the social relationships extracted from short-range wireless networks and mapped to online social network.We use an objective evaluation criterion for the verification and feedback steps of the proposed framework to verify the performance of the framework.Different from other models,the proposed algorithm has reference data for us to judge the performance of the algorithm,and these algorithms are universal for this type of data.If the research result does not reach the expected value when using a certain algorithm for a certain data set,it does not mean that the algorithm is wrong.In this case,first replace the data indicator used for the algorithm.If the result is still not ideal,it can only indicate that the algorithm is not suitable for processing this data set and then replace other algorithm research.This paper innovatively studies two types of networks and proposed multiple algorithms.And verify these algorithms with five sets of data sets.The final results show the feasibility and effectiveness of our proposed framework model.