Abstract
The smart home has begun playing an important role in supporting independent living by monitoring the activities of daily living, typically for the elderly who live alone. Activity recognition in smart homes has been studied by many researchers with much effort spent on modeling user activities to predict behaviors. Most people, when performing their daily activities, interact with multiple objects both in space and through time. The interactions between user and objects in the home can provide rich contextual information in interpreting human activity. This paper shows the importance of spatial and temporal information for reasoning in smart homes and demonstrates how such information is represented for activity recognition. Evaluation was conducted on three publicly available smart-home datasets. Our method achieved an average recognition accuracy of more than 81% when predicting user activities given the spatial and temporal information.
Highlights
Almost every country in the world is experiencing a growing and aging population
This paper extends the prediction by partial matching (PPM), an adaptive statistical data compression technique, to include spatial and temporal information
This paper shows the importance of spatial and temporal information in interpreting human activity and how such information can be represented for activity recognition
Summary
The smart home is considered a viable solution to address living problems, typically the elderly or those with diminished cognitive capabilities. An important part of the functioning of smart homes is to monitor user daily activities and detect any alarming situations (e.g., skipping meals several days in a row). Sensors attached to objects of daily use (e.g., fridge, light, etc.) are often deployed in the smart home to collect information about user daily activities. These sensors are activated when the user performs their activities (e.g., opening the fridge, turning on the light, etc.). The recognition system uses the sensory outputs from the home to learn about user activity patterns and predict the probable event
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
