Paper Title
The Impacts of Internet Anonymity on Group Discussion

Abstract
Recently social media have become a valuable resource of text mining. Research on various types of social network has gained much attention recent years. However, far too little attention has been paid to instant messaging, which is a format of computer-mediated communication (CMC) and contains a huge amount of text data. Laboratory experiment method approach was employed to build text corpus. Participants of the experiment attended two different discussion environments: anonymous and named. Subject in each discussion environment would discuss a specific social issue. A popular corpus named NTUSD was employed to analyze the messages posted in different social context. A corpus is a collection of written or spoken material in machine-readable form, assembled for the purpose of linguistic research. After words segmentation, words were marked as a sentiment word by using sentiment dictionary “NTUSD” and given a sentiment score by linking WordNet and SentiWordNet. This study use sentiment score to indicate the degree of emotional level of each individual who participated in the social issue discussion. Higher sentiment score indicates more emotional during the discussion. The results showed that anonymity have significant effects on the sentiment scores. In conclusion, anonymous environment will more likely to make online discussants more emotional and may result in a more frank discussion. Keywords - Computer-mediated Communication; Text Mining; Sentiment Analysis; Internet Anonymity.