Forecasting Word Model: Japanese Simplification for Non-Native Speakers
Muhaimin Hading (1551204)
This thesis introduces Japanese lexical simplification. Japanese lexical simplification is the task of replacing difficult words in a given Japanese sentences to produce new sentences with simple words without changing the original meaning of the sentences. We purposed a method of supervised regression learning to estimate the difficulty ordering of words with statistical features obtained from two types of Japanese corpora. For the similarity of the words, we used a Japanese thesaurus, Japanese lexical simplification system (SNOW S3), and dependency based word embeddings. We conducted two types of evaluation; (1) evaluation of the word difficulty is performed by comparing the difficulty ordering of the words, and (2) evaluation of new simplified sentences by human judge.