TY - JOUR

T1 - Rumor conversations detection in twitter through extraction of structural features

AU - Lotfi, Serveh

AU - Mirzarezaee, Mitra

AU - Hosseinzadeh, Mehdi

AU - Seydi, Vahid

PY - 2021/12

Y1 - 2021/12

N2 - Twitter is one of the most popular and renowned online social networks spreading information which although dependable could lead to spreading improbable and misleading rumors causing irreversible damage to individuals and society. In the present paper, a novel approach for detecting rumor-based conversations of various world events such as real-world emergencies and breaking news on Twitter is investigated. In this study, three aspects of information dissemination including linguistic style used to express rumors, characteristics of people involved in propagating information and structural features are studied. Structural features include features of reply tree and user graph. Structural features were extracted as new features in order to enhance the efficiency of the rumor conversations detection. These features provide valuable clues on how a source tweet is transmitted and responds over time. Experimental results indicate that the new features are effective in detecting rumors and that the proposed method is better than other methods as F1-score increased by 4%. Implementation of the proposed method was carried out on Twitter datasets collected during five breaking news stories.

AB - Twitter is one of the most popular and renowned online social networks spreading information which although dependable could lead to spreading improbable and misleading rumors causing irreversible damage to individuals and society. In the present paper, a novel approach for detecting rumor-based conversations of various world events such as real-world emergencies and breaking news on Twitter is investigated. In this study, three aspects of information dissemination including linguistic style used to express rumors, characteristics of people involved in propagating information and structural features are studied. Structural features include features of reply tree and user graph. Structural features were extracted as new features in order to enhance the efficiency of the rumor conversations detection. These features provide valuable clues on how a source tweet is transmitted and responds over time. Experimental results indicate that the new features are effective in detecting rumors and that the proposed method is better than other methods as F1-score increased by 4%. Implementation of the proposed method was carried out on Twitter datasets collected during five breaking news stories.

U2 - 10.1007/s10799-021-00335-7

DO - 10.1007/s10799-021-00335-7

M3 - Article

VL - 22

SP - 265

EP - 279

JO - Information Technology and Management

JF - Information Technology and Management

SN - 1573-7667

IS - 4

ER -