Recognizing Context-Aware Human Sociability Patterns Using Pervasive Monitoring for Supporting Mental Health Professionals

Abstract

Traditionally, mental health specialists monitor their patients’ social behavior by applying subjective self-report questionnaires in face-to-face meetings. Usually, the application of the self-report questionnaire is limited by cognitive biases (e.g., memory bias and social desirability). As an alternative, we present a solution to detect context-aware sociability patterns and behavioral changes based on social situations inferred from ubiquitous device data. This solution does not focus on the diagnosis of mental states, but works on identifying situations of interest to specialized professionals. The proposed solution consists of an algorithm based on frequent pattern mining and complex event processing to detect periods of the day in which the individual usually socializes. Social routine recognition is performed under different context conditions to differentiate abnormal social behaviors from the variation of usual social habits. The proposed solution also can detect abnormal behavior and routine changes. This solution uses fuzzy logic to model the knowledge of the mental health specialist necessary to identify the occurrence of behavioral change. Evaluation results show that the prediction performance of the identified context-aware sociability patterns has strong positive relation (Pearson’s correlation coefficient >70%) with individuals’ social routine. Finally, the evaluation conducted recognized that the proposed solution leading to the identification of abnormal social behaviors and social routine changes consistently.

Sensing Apps and Public Data Sets for Digital Phenotyping of Mental Health: Systematic Review

“…Researchers also used identified behaviors to design ML models capable of classifying and predicting mental states [ 32 , 80 ], which can be used as decision support tools for health professionals. Lastly, some studies [ 31 , 81 , 82 ] did not report on additional analyses, but concentrated on describing the features of their sensing apps to facilitate DPMH research.…”

“…Morshed et al [ 81 ] developed a computational method to predict mood stability from behavioral features (eg, frequency of conversation, number of location changes, and duration of different physical activities) extracted from accelerometer, microphone, GPS, and Wi-Fi. Recently, de Moura et al [ 82 , 83 ] developed a solution capable of detecting sociability patterns and routine changes in social event streams (ie, conversation events).…”

“…This is crucial in extreme cases such as signs of suicidal ideation, but generally useful where the goal is to implement ecological momentary interventions or just-in-time interventions that rely on just-in-time information on user status. In this sense, both rule-based engines (eg, fuzzy logic [ 39 ], complex event processing [ 82 ]) and ML-based approaches [ 20 ] are promising tools to process context data efficiently and infer high-level information in DPMH.…”

Sensor Networks: Physical and Social Sensing in the IoT

“…The article entitled “Recognizing Context-Aware Human Sociability Patterns Using Pervasive Monitoring for Supporting Mental Health Professionals”, contributed by Moura et al [ 8 ], presents a proposal to detect context-aware sociability patterns. This would enable the identification of patterns in the periods of day in which users socialize, while also supporting the detection of abnormal behaviour and changes in daily routine.…”