What is it about?
Data coming from sensors in smart homes allows us to recognize daily activities such as eating, taking a shower, cooking. By analyzing these long-term data we can discover the routines of people at home and identify when these routines change. We focus in temporal patterns, such as: - what is the usual time for breakfast? By analyzing the temporal interval we can also analyze how consistent the routine is through the days. We also include the context to avoid false alarms. We can discover weekend routines vs week routines, or days that are different, or analyze if rain changes the routines.
Featured Image
Why is it important?
Changes in routines can be an indicator of health decline or cognitive health decline so identifying them is important. Context analysis serves two purposes: avoiding false alarms by changes that are frequent on a single context and discovering context clues that make a behavior change. We consider privacy by processing the data as it comes, not storing it. We keep the routines in a local server (Edge) so only notifications are sent to the care network of the user.
Read the Original
This page is a summary of: Learning and managing context enriched behavior patterns in smart homes, Future Generation Computer Systems, February 2019, Elsevier,
DOI: 10.1016/j.future.2018.09.004.
You can read the full text:
Contributors
The following have contributed to this page
