This paper presents an automatic acquisition of newly Myanmar opinion words based on syllable n-gram frequencies because Myanmar is as syllabic script and Myanmar language uses a syllabic writing system. As there is a need for automatically constructing opinion lexicons for any specific domain, the proposed method makes it possible to construct opinion lexicon for Myanmar movie domain in a bootstrapping manner. We use Myanmar movie comments from popular social network Facebook as a corpus and determine the valence or polarity of the comments, whether it is positive or negative by using a small amount of opinion words. Then opinion words are extracted from polarity-identified Facebook comments. Our proposed method is based on n-gram of syllables without word segmentation, since Myanmar is less-resourced language and there is no public tool for word segmentation.
@InProceedings{THANT18.1, author = {Win Win Thant and Kiyoaki Shirai}, title = {Automatic Acquisition of Opinion Words from Myanmar Facebook Movie Comments}, booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)}, year = {2018}, month = {may}, date = {7-12}, location = {Miyazaki, Japan}, editor = {Kiyoaki Shirai}, publisher = {European Language Resources Association (ELRA)}, address = {Paris, France}, isbn = {979-10-95546-24-5}, language = {english} }
Summary of the paper