2026-04-30T21:13:51Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/00006746 2021-12-13T05:08:55Z 1582963342780:1596102355557 user-uit

Automatic Acquisition of Opinion Words from Myanmar Facebook Movie Comments Win Win Thant Kiyoaki Shirai This paper presents a method for the automatic acquisition of new Myanmar opinion words based on syllable n-gram frequencies, since the Myanmar language uses a syllabic writing system. There is a need for the automatic construction of opinion lexicons for any specific domain, and the proposed method makes it possible to construct an opinion lexicon for Myanmar movies using a bootstrapping approach. We use Myanmar movie comments from Facebook, a popular social network, as a corpus, and determine the valence or polarity of the comments (positive or negative) using a small number of opinion words. These opinion words are then extracted from polarity-identified Facebook comments. Our proposed method is based on n-grams of syllables without word segmentation, since Myanmar is an under-resourced language and no public tool exists for word segmentation. 2018-05-08 https://meral.edu.mm/records/6746