User Attribute Information Inference Method Based On Strong Relation

At present,the Internet era has developed rapidly,and more and more users have begun to enter the new social age-the network social age,which makes the online social network develop rapidly.Social platforms such as weibo,We Chat,QQ and renren have emerged one after another and gradually have a large number of users.Users in social networking platform to fill out your personal information,including the user's gender,age,geographic information,hobbies,family background and other attribute information,makes the information platform for managers and researchers in social network can better research management.However,in the current user attribute information,there is often a phenomenon of lack of information or false information.The imperfect user data in the network will affect the research of social network and the operation of the platform.In order to solve this phenomenon,a more perfect information inference method for unknown attributes of users is proposed.However,the results are good and the classical information inference method does not verify the comparative inference effect on the same large real social network data platform.At the same time,some of the research are carried out on the unauthorized network.It does not take into account the influence of the weight difference between nodes on the inference of unknown attribute information.This topic is mainly based on the problem of user missing information in the social network at the present stage.In the same large-scale social network data set analysis and experiments,the effect of the present classical user attribute information inference method is verified,and the user information of the unknown attribute is pushed for the first time according to the relationship strength between the nodes.In essence,breaking the improved idea is to remove the noise data in the network by using the strength of the relationship between nodes.Using the knowledge of the strong and weak relationships in the social network,the original social network data can be converted into the research work of information inference for the weighted connected data network tagged the strong and weak edges.In the inference experiment,we also verify that the relationship strength between nodes will have a positive impact on the inference of regional information of unknown attribute nodes.At the same time,the relationship between users' strength and weakness has obvious difference in the effect of inferential improvement between different attributes.In the experiment,it is proved that using the relationship intensity to remove thenoise data can be more efficient to infer the false information and the missing information in the social network.The strong relationship is meaningful to the users,the groups and the network of the social network.We need to make rational use of strong ties to provide better research methods and methods for user information in social networks.