Dr. Olga Streibel
Can we make use of knowledge and time
in data-based situation recognition?
Vortrag zum Statusmeeting vom 7. Oktober 2014
This talk introduces the vision of the TreMiSR project, currently being conducted at the National Institute of Informatics in Tokyo, Japan. The goal of the Trend Mining for Situation Recognition project is to create intelligent, context-aware methods for situation recognition. One of its particular focuses lies in the sensor-free technology that aims in assisting elderly people in their comfortable living, despite the handicaps that one experiences being elderly. In general, trend mining with its trend template approach support the extraction of implicit, previously unknown and potentially useful information from time-ordered texts or data. The trend template as a knowledge-based approach for mining trends assumes that mining with knowledge helps us in understanding a trend. Understanding a trend brings new solutions into the problems of situation recognition.