A Character Similarity Characterization Method Based on Heterogeneous Data

Abstract

The invention discloses a character similarity characterization method based on heterogeneous data, which belongs to the field of data mining. The present invention firstly collects microblog texts of users, obtains the attention relationship between users and the basic information of each user, and selects a processing method individually according to the characteristics of different types of data, and adopts the Doc2vec model for microblog texts, combines the context information to convert the text Express it as a vector, then measure the similarity according to the defined similarity function, and finally fuse the matrices obtained from different dimensions to describe the final similarity of users. The present invention introduces a variety of social network information, including social relationship data, user attribute data, and user text data, etc., and by comprehensively considering different types of information, a more comprehensive character similarity characterization method is obtained; at the same time, the present invention provides For a variety of data processing and calculation schemes, use complete data and weighted fusion methods to personalize and calculate the similarity of people with different preferences.

Description

technical field [0001] The invention belongs to the field of data mining and relates to a similarity calculation technology, in particular to a character similarity characterization method based on heterogeneous data. Background technique [0002] With the development of the Internet, people's lives are increasingly inseparable from the Internet, relying on the Internet to work, socialize and express their opinions more and more closely, resulting in the gradual blurring of the boundaries between online and offline. During this process, people express their interests and characteristics through different forms; however, in the face of a huge amount of information, people have to identify the content they are interested in and find friends with the same interests, businesses and governments from the massive amount of information. It is becoming more and more difficult for institutions to find user groups that are more valuable to them, and to conduct further research or recom...

AI Technical Summary

Problems solved by technology

During this process, people express their interests and characteristics in different forms; however, in the face of a huge amount of information, people have to identify the content they are interested in and find like-minded friends, businesses and governments from the massive amount of information. It is becoming more and more difficult for institutions to find more valuable user groups and conduct further research or recommendations based on user information

Images