Reviewers Classification in an Online Community of Romanian Tourists

Our previous work addressed the computational analysis of communities of Romanian online users involved in tourism activities and interested in sharing their impressions and experiences, by focusing on touristic content sharing and review sites. Our studies comprise several tasks such as sentiment analysis, keywords extraction and graph-based structuring of the users community. In this paper we extend our previous analysis of the users community of AmFostAcolo tourism Web site, by proposing a supervised classification method of reviewers based on their portfolio data. The goal of our classification is to develop a method for automated labeling the users that post reviews as novice, common or experienced reviewers. The classification uses features derived from data extracted from users’ portfolio. We have investigated several multi-class classifications and our results are encouraging.