Sentiment analysis in Indonesian and French by SentiSAIL

SentiSAIL is a multilingual, lexicon-based sentiment analysis software tool. The implementation of a general methodology allows the software core to remain unchanged, while supporting the processing of new (previously unsupported) languages through a framework of language- specific features. Support for new languages or new domains can be added in a short period of time, reducing development efforts and costs. The current work presents the development and evaluation of two new languages, belonging to different language families: Indonesian (Bahasa Indonesia), a member of the Malayo-Polynesian language family and French, a member of the Indo-European language family. Evaluation on a self- compiled and annotated traditional media corpus shows the performance of SentiSAIL to be sufficiently high and comparable to the average rate of native speakers.