User Performance Prediction Based on Their Behavioural Factors

Predicting user performance based on behavioural factors involves analysing users’ actions, interactions, and historical patterns to anticipate their future behaviour and performance. This approach utilizes data on user engagement, preferences, and past activities to enhance personalized user experiences, optimize system functionality, and forecast user actions for better decision-making across various domains. With the advancement of web and wireless technologies, users can access services through mobile devices anytime and anywhere, leading to the proliferation of health-related mobile apps. Despite their availability, the adoption of these apps for health and care has been uneven. Understanding mobile user behaviour is pivotal for enhancing system performance and Quality of Service (QoS). This paper presents a robust framework for mining and predicting user behaviour patterns, comprising three phases. The model integrates a similarity inference model to gauge similarity among stores, items, and pages. The proposed methodology utilizes the Personal Mobile Health Pattern Mine (PMHP-Mine) algorithm and GIT-Tree for uncovering mobile user behaviour patterns, while prediction is facilitated by the Longest Chain Subsequence algorithm. The evaluation demonstrates efficient performance results.