Abstract
Assessing the semantic similarity of texts is a fundamental concept which has many applications in natural language processing and related fields. This work presents both word and sentence semantic similarity measures specifically for Thai language. The word similarity measure is based on word embedding vectors, WordNet database and an edit-distance measure. The sentence similarity measure relies on the word similarity measure as a baseline. The proposed measures are compared with existing methods on benchmark datasets.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have