Machine Learning & Knowledge Extraction for Ambient Assisted Living (AAL*)
*) AAL includes all aspects for supporting our elderly generation to live a healty, safe and active life
Workshop organized by Johannes KROPF (AIT Austrian Institute of Technology, AT), Liming CHEN (DeMontfort University, UK) & Parisa RASHIDI (University of Florida, US)
The session aims to bring together care professionals with machine learning experts.
Ambient assisted living (AAL) can be defined as “the use of information and communication technologies (ICT) in a person’s daily living and working environment to enable them to stay active longer, remain socially connected and live independently into old age” ( www.aal-europe.eu ).
Research in the AAL community covers a wide range of topics, but one of the largest is human activity recognition and behavior understanding. The main objective is the detection of human activity, actions and situations in smart environments and/or based on wearable sensors or smart devices. A huge interest lies in the detection of falls, critical situations and changes in long term behavior (e.g. early detection of neurodegenerative diseases). This calls for data integration, data fusion and machine learnign and knowledge extraction technologies!
Even there are overlaps with other applications when it comes to methods used, AAL as its own challenges: human behavior differs very much individually and the data gained from smart home sensors are sparse and uncertain. Moreover it is very difficult to gain enough training data for standard models (such as HMM), hence, more sophisticated and/or unsupervised methods are required. On the other hand, wearables and smart watches are appearing on the market which are equipped with more and more sensors which could be used.
In addition to application papers, manuscripts dealing with fundamental questions and theoretical aspects in machine learning, as needed for successful applications in AAL, are encouraged.
Research topics covered by this special session include but are not limited to the following topics:
- Behaviour recognition
- Fall detection
- Emotional computing
- Human-computer interaction
- Smart home technologies
Accepted Papers will be published in the Springer CD-MAKE Volume of Lecture Notes in Artificial Intelligence (LNAI). We are planning to invite outstanding contributions for extensions in journals (tba.)
For submission details please proceed to the CD-MAKE authors area