Technical Factors in Health App Adoption: Analyzing User Responses with Long Short-Term Memory Algorithm

In Indonesia, health applications have emerged as a crucial component in improving healthcare access and quality. Despite the rapid growth of health applications, their adoption remains suboptimal due to challenges such as limited digital infrastructure, varying levels of IT literacy, and concerns about data privacy. This study seeks to address these challenges by categorizing the acceptability of health applications based on technical factors, utilizing the Long Short-Term Memory (LSTM) algorithm. The research is based on data collected through interviews with health application users in Bandung Regency with a primary focus on parents with children. The study compares two LSTM model scenarios: single-layer and double-layer. The findings reveal that perceived usefulness is the most significant factor influencing positive user responses, with system quality and facilitating conditions also playing crucial roles in enhancing user acceptance. The single-layer LSTM model achieved the highest accuracy of 84.5%, compared to 83.5% for the double-layer model. This study offers valuable strategies for developers to refine application features, ultimately contributing to improved healthcare accessibility and outcomes in Indonesia.