Automatic Test Cases Generation Method by Structuring Specification Documents Using Seq2Seq
Yuki Shimizu, K. Ueda · 2025
We aim to automate the development of communication software. By leveraging machine learning, we propose a method for automatically generating system testing test cases from requirement specification documents. In this research, we focus on developing a method to generate test cases by structuring requirement specification sentences using a deep learning Seq2Seq model with an Attention mechanism. Our goal is to further enhance the accuracy of this approach. To evaluate the effectiveness and characteristics of the proposed method, we conducted experiments using requirement specifications of communication system software and various IT system requirement specification documents.