[1J1] [General Session] 9. NLP / IR

[1J1-03] Automatic generation of questions from product manual sentences

〇Sato Sato¹, Itsui Hiroyasu², Okumura Manabu³ (1. School of Engineering, Tokyo Institute of Technology, 2. Mitsubishi Electric Corporation, 3. Institute of Innovative Research, Tokyo Institute of Technology)

Keywords: Natural language processing, Question generation

In this research, as the first step to develop a system that automatically generates related question answer pairs from sentences in a product manual, we present a method to automatically generate question sentences from manual sentences. As the result of experiments with a data set of about 1400 sentences, we obtained BLEU score of 62.11 by comparing generated sentences with manually prepared question sentences.