Who Leaves Malicious Comments on Online News? An Empirical Study in Korea

Abstract

Because online news comments have a strong influence on the reader's perception of public opinion, there is a call for efforts to reduce the adverse impact of online news comments, particularly malicious ones. Although many online news platforms currently use technology to detect malicious comments automatically, there is still a technical limit in identifying malicious comments. To improve detection accuracy, it is necessary to understand not only malicious comments but also malicious commenters. Despite the importance of understanding malicious commenters, there is little empirical research on their characteristics. This study aims to understand the characteristics of malicious commenters and develop a prediction model based on their features using real data of users and commenting activities from Naver, a leading Internet news portal in Korea. This study found that the demographic characteristics of malicious commenters tend to be those of males and older people. In terms of commenting activities, the online news commenters who leave more comments per news article and per day, delete more comments, and leave longer comments tend to be malicious commenters.