KoEPT 기반 한국어 수학 문장제 문제 데이터 분류 난도 분석 = Analyzing Korean Math Word Problem Data Classification Difficulty Level Using the KoEPT Model

다국어 초록 (Multilingual Abstract)

In this paper, we propose KoEPT, a Transformer-based generative model for automatic math word problems solving. A math wordproblem written in human language which describes everyday situations in a mathematical form. Math word problem solving requiresan artificial intelligence model to understand the implied logic within the problem. Therefore, it is being studied variously across theworld to improve the language understanding ability of artificial intelligence. In the case of the Korean language, studies so far havemainly attempted to solve problems by classifying them into templates, but there is a limitation in that these techniques are difficultto apply to datasets with high classification difficulty. To solve this problem, this paper used the KoEPT model which uses ‘expression’tokens and pointer networks. To measure the performance of this model, the classification difficulty scores of IL, CC, and ALG514, whichare existing Korean mathematical sentence problem datasets, were measured, and then the performance of KoEPT was evaluated using5-fold cross-validation. For the Korean datasets used for evaluation, KoEPT obtained the state-of-the-art(SOTA) performance with 99.1%in CC, which is comparable to the existing SOTA performance, and 89.3% and 80.5% in IL and ALG514, respectively. In addition, asa result of evaluation, KoEPT showed a relatively improved performance for datasets with high classification difficulty. Through an ablationstudy, we uncovered that the use of the ‘expression’ tokens and pointer networks contributed to KoEPT’s state of being less affectedby classification difficulty while obtaining good performance.