In recent years, activity recognition in smart homes is an active research area due to its applicability in many applications, such as assistive living and healthcare. Besides activity recognition, the information collected from smart homes has great potential for other application domains like lifestyle analysis, security and surveillance, and interaction monitoring. Therefore, discovery of users common behaviors and prediction of future actions from past behaviors become an important step towards allowing an environment to provide personalized service. In this paper, we develop a unified framework for activity recognition-based behavior analysis and action prediction. For this purpose, first we propose kernel fusion method for accurate activity recognition and then identify the significant sequential behaviors of inhabitants from recognized activities of their daily routines. Moreover, behaviors patterns are further utilized to predict the future actions from past activities. To evaluate the proposed framework, we performed experiments on two real datasets. The results show a remarkable improvement of 13.82% in the accuracy on average of recognized activities along with the extraction of significant behavioral patterns and precise activity predictions with 6.76% increase in F-measure. All this collectively help in understanding the users” actions to gain knowledge about their habits and preferences.

The identification of a mental disorder at its early stages is a challenging task because it requires clinical interventions that may not be feasible in many cases. Social media such as online communities and blog posts have shown some promising features to help detect and characterise mental disorder at an early stage. In this work, we make use of user-generated content to identify depression and further characterise its degree of severity. We used the user-generated post contents and its associated mood tag to understand and differentiate the linguistic style and sentiments of the user content. We applied machine learning and statistical analysis methods to discriminate the depressive posts and communities from non-depressive ones. The depression degree of a depressed post is identified using variations of valence values based on the mood tag. The proposed methodology achieved 90%, 95% and 92% accuracy for the classification of depressive posts, depressive communities and depression degree, respectively.