返回
Assessing topic-based users credibility in twitter
DOI:10.1007/s11042-023-18093-9.png)
摘要
En 中文
Online Social Networks (OSN) have become an inevitable source of information. Every user in OSNs can share true or false information regardless of their knowledge. False information can cause damage to people, companies, and even societies. Thus investigating the correctness of information on OSN is crucial. Several researchers worked on assessing users' credibility because if a person is considered credible, so will the information he/she shares. In this paper, we introduce a novel approach that assesses topic-based credibility of users where a user's credibility varies with topics. First, we assess the topic-based credibility of a user's tweets and then calculate users' expertise in a range of topics. Afterward, we introduce new graph-based measures that consider semantic and structural aspects to assess the influence of experts versus the influence of rumor spreaders on the user's credibility. Finally, in the experimental section, the impact of working on a topic basis on tweets' credibility is investigated showing that topic-based results are better than topic-independent tweets' credibility results. In addition, the topic-based credibility of OSN users and how it is influenced by experts and rumor spreaders is analyzed revealing that experts have a positive and strong impact compared to rumor spreaders' negative impact on users' credibility results.
Keyword:
Credibility
Online social networks
Natural language processing
Semi-automatic annotation
Twitter
期刊
IF:
3
论文数:
2.0W
被引数:
3.2W
机构
引用论文
Comparison of AMSR-2 wind speed and sea surface temperature with moored buoy observations over the Northern Indian OceanAMSR-2风速和海表温度与北印度洋系泊浮标观测的比较
Seasonal rainfall variability in the southern Mediterranean border: Observations, regional model simulations and future climate projections南地中海边界区域的季节性降雨变率:观测、区域模型模拟和未来气候预测
Atmósfera
IF0

