AICCSA : AI tool for collecting and classifying sports applications

Mobile applications for sports and, Health & Fitness form a vast and rapidly growing ecosystem, where comprehensive research requires access to extensive application datasets. However, collecting and categorizing such data for Sport Human-Computer Interaction research using manual methods is a highly labour-intensive and time-consuming process. To address this challenge, this thesis adopts a Design Science Research methodology to develop and evaluate an AI-based software tool called AI tool for Collecting and Classifying Sports Applications (AICCSA) that automates the collection and categorization of mobile application data. Recent advances in generative artificial intelligence have enabled the development of tools capable of automating repetitive tasks in areas such as data processing, programming, and knowledge work. The proposed tool utilizes publicly available APIs for data collection and leverages OpenAI’s GPT-5-mini model to classify applications by sport type, purpose, and primary user group. The performance of the proposed artifact was assessed by comparing its classification outcomes with those produced by a human evaluator. The results demonstrate a high level of potential for using AI-based software tools to support Sport HCI research by enabling efficient collection and classification of large application datasets.