Authorship Attribution in Greek Tweets Using Author's Multilevel N-Gram Profiles

George K. Mikros Department of Italian Language and Literature National and Kapodistrian University of Athens Email: [email protected] Website: http://users.uoa.gr/~gmikros The aim of this study is to explore authorship attribution methods in Greek tweets. We have developed the first Modern Greek Twitter corpus (GTC) consisted of 12,973 tweets crawled from 10 Greek popular users. We used this corpus in order to study the effectiveness of a specific document representation called Author’s Multilevel N-gram Profile (AMNP) and the impact of different methods on training data construction for the task of authorship attribution. In order to address the above research questions we used GTC to create 4 different datasets which contained merged tweets in texts of different sizes (100, 75, 50 and 25 words). Results were evaluated using authorship attribution accuracy both in 10-fold cross-validation and in an external test set compiled from actual tweets. AMNP representation achieved significant better accuracies than single feature groups across all text sizes. A second experiment was conducted in order to evaluate whether AMNP representation captures better the stylometric profile of the tweets than using separate n-gram profiles. For this reason we repeated the authorship attribution task in the four datasets of varying text size chunks under both testing conditions (10-fold cv and external tweets). The results appear to the following figures (figure 3 & 4).