Integrating Mobile Devices with Cohort Analysis into Personalised Weather-Based Healthcare.

Mobile healthcare applications can empower users to self-monitor their health conditions without the need to visit any medical centre. However, the lack of attention on engagement aspects of mobile healthcare applications often result in users choosing to uninstall the application after the first usage experience. This results in failure of effective prolonged personalised healthcare, especially for users with chronic disease related to weather conditions such as asthma and eczema which require long-term monitoring and self-care. Therefore, this paper aims to identify the pattern of application user engagement with a weather-based mobile healthcare application through cohort retention analysis. Enhancement features for improving the engagement of personalised healthcare can provide meaningful insight. The proposed application allows the patient to conduct disease control tests to check the severity of their condition on a daily basis. To measure the application engagement, we distribute the mobile application designed for primary testing over a period of ten days. Based on the primary testing, data related to retention rate and the number of control test reported were collected via Firebase Analytic to determine the application engagement. Subsequently, we apply cohort analysis using a machine learning clustering technique implemented in Python to identify the pattern of the engagement by application users. Finally, useful insights were analysed and implemented as enhancement features within the application for improving the personalised weather-based mobile healthcare. The findings in this paper can assist machine learning facilitators design effective use policies for weather-based mobile healthcare with fundamental knowledge enhanced with personalisation and user engagement.