Dialogue Act Recognition for Text-based Sinhala

This paper discusses the application of classical machine learning approaches to the task of Dialogue Act Recognition for text-based Sinhala. A study was carried out to identify a dialogue act tag set for Sinhala. A new corpus using Sinhala subtitles for English movies was created and was annotated with the selected dialogue acts. Evaluation of the dialogue act recognition system was performed using features that were used for English language, plus the newly identified features for Sinhala. Although Sinhala is an under-resourced language without even the basic tools such as a PoS tagger, we managed to achieve good classification accuracy by exploiting Sinhala specific features. As far as we are aware, this is the first research on dialogue act recognition on the family of Indo-Iranian languages.