Establishment of an identification model and identification method for continuous followers of events in Weibo

Abstract

The invention discloses a method for establishing an identification model of a continuous follower of an event in a microblog, the method comprising: establishing a training sample set, including users following a certain event and microblogs related to the event; extracting training samples respectively Concentrating each user's attention degree feature and recognition degree feature, respectively training attention degree user classification model and ranking model, the identification model of the continuous follower of the event in the microblog includes the trained attention degree user classification model and ranking model. Based on the above recognition model, the present invention also provides a method for identifying persistent followers of events in microblogs, which can identify persistent followers of events, and obtain relatively dense, complete The collection of event information can effectively improve the efficiency of specific information retrieval.

Description

technical field [0001] The present invention relates to the technical field of information retrieval, in particular to the technical field of expert information retrieval, and in particular to an identification model establishment and identification method for continuous followers of events in microblogs. Background technique [0002] The content of Weibo is short and the related information is highly scattered, which has brought impact and test to the way of information acquisition and retrieval. Using traditional information retrieval methods to track specific events or topics in microblogs is strongly dependent on the choice of search terms. In the context of massive microblog information, it is often faced with useful data sparseness, data redundancy, topic drift and the derivation of sub-events. The resulting incomplete retrieval information and many other difficulties. A notable feature of the new media environment is to organize information with the user as the core....

AI Technical Summary

Problems solved by technology

Using traditional information retrieval methods to track specific events or topics in microblogs is strongly dependent on the choice of search terms. In the context of massive microblog information, it is often faced with useful data sparseness, data redundancy, topic drift and the derivation of sub-events. Many difficulties caused by incomplete retrieval information

Images