The Application of User Event Log Data for Mental Health and Wellbeing Analysis

Abstract

Many digital interaction technologies, including web-based interventions, smartphone applications, and telephone helplines, can provide a basis for capturing real time data of interactions between the user and the system. Such data is recorded in the form of log files, which records user events that range from simple keystrokes on a computer, user activated sensor data or duration/frequency of phone calls. These interactions can provide rich datasets amenable to user data analytics using machine learning and other analytics techniques. This data analysis can highlight usage patterns and user behaviours based on their interaction with the technology. User log data analysis can be descriptive statistics (what users have done), predictive analytics (what events will happen) and prescriptive (what action to take given a predicted event or outcome). This can also be thought of as spanning across different levels of user analytics from hindsight, insight and foresight. Predictive analytics are used with log data to provide predictions on future user behaviour based on early usage behaviours. Event logs are objective regarding usage, but usage may not correlate with the level of the system’s user experience. Hence, ecological momentary assessment (EMA) of the user experience can be used augment user log data. Nevertheless, with the emergence of health applications and other app-based health services, we consider how user event logs can be specifically used within the mental health domain. This can provide beneficial insights into how users interact with mental health e-services, which can provide an indication of their current and future mental state.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.