Comparing Frequency- and Style-Based Features for Twitter Author Identification
Rachel M. Green · 2016
Author identification is a subfield of Natural Language Processing (NLP) that uses machine learning tech-niques to identify the author of a text. Most previous research focused on long texts with the assumption that a minimum text length threshold exists under which au-thor identification would no longer be effective. This pa-per examines author identification in short texts far be-low this threshold, focusing on messages retrieved from Twitter (maximum length: 140 characters) to determine the most effective feature set for author identification. Both Bag-of-Words (BOW) and Style Marker feature sets were extracted and evaluated through a series of 15 experiments involving up to 12 authors with large and small dataset sizes. Support Vector Machines (SVM) were used for all experiments. Our results achieve clas-sification accuracies approaching that of longer texts, even for small dataset sizes of 60 training instances per author. Style Marker feature sets were found to be sig-nificantly more useful than BOW feature sets as well as orders of magnitude faster, and are therefore suggested for potential applications in future research.