Adaptable Expression Search Framework with Customizable Pattern Matching for Language Studies

This study introduces a novel design for a pattern matching system capable of extracting select words or phrases from texts.In the process of learning a foreign language, searching for instances of usage or grammatical structures within texts is a common requirement.While numerous systems, particularly concordancers, have been proposed in prior research, many of them lacked flexibility and posed challenges when attempting to combine specific search patterns.To address this limitation, we developed a new phrase search system that allows users to craft their search patterns by merging basic search templates.This paper presents a system that leverages Prolog predicates as a fundamental data structure, utilizing SWI-Prolog for processing.The system is capable of performing searches that integrate regular expressions with other combined patterns.Our performance test demonstrates the system can process 10,000 sentences without errors.User evaluation employing system usability scale indicates that while the current usability of our system requires enhancement, the feedback gathered from these evaluations not only confirms the system's robustness but also provides valuable insights for future improvements.