Detecting of topic-specific leaders in social networks

Abstract:

Social networks on the web are growing dramatically in size and number. Identifying important people such as leaders in social networks have attracted the attention of researchers and the media. Approaches to perform this task are based just on the network topology and non-topic specific. We propose LeadershipRank, a topic-specific leadership score for detection of leaders. A Markov model is built to calculate importance inside a social network. It is based on sociological theories and graph theory to provide an accurate measure. We evaluate the performance of the algorithm in a real-world social network such as Twitter against traditional metrics.

Año de publicación:

2019

Keywords:

• Data Mining
• Machine learning
• SOCIAL BEHAVIOR

Fuente:

scopus