Supporting Test Case Design on Reasoning Scheme with Natural Language Processing Technique
Noriyuki Kushiro, Yusuke Ogata · 2021 IEEE International Conference on Big Data (Big Data) · 2021
The purpose of the study is to support test case design by introducing natural language processing techniques and reasoning schemes for designing. Sentences containing a logical relationship of implication (key sentences) are extracted from a specification document with natural language processing techniques, and supplementary explanation sentences for the key sentence (subordinate sentences) that complement parameters appeared in the key sentence are placed in Toulmin's reasoning scheme based on "similarity" to the key sentence. As a result, we succeeded in automatically generating test cases for product testing, and also in supporting to extract parameters for each test case.