A Framework of Web-Based Social Network Search System

Social network mining technology and its implementation are the subject of much interest and difficulty in recent research. A framework of web-based social network search system is designed to carry out social network search. The paper presents a method to implement web downloader in order to obtain web documents containing social network information. Characteristic vector weight is calculated with C-value and Inverse Document Frequency, and identical name judgment is carried out by cataloguing web documents with clustering algorithm. The paper also gives a definition of the strength of personal relation, and current commercial search engines are used to design and implement algorithms that can retrieve personal relation information and calculate the strength of those relations. We tested the method on a search engine for name search, and the results show that identical name judgement based on a modified hierarchical clustering algorithm can significantly improve performance.