Title :
Routine learning: analyzing your whereabouts
Author :
Pirttikangas, Susanna ; Riekki, Jukka ; Röning, Juha
Author_Institution :
Dept. of Electr. & Inf. Eng., Oulu Univ., Finland
Abstract :
A routine is a temporal context sequence that occurs often. In routine learning, already recognized contexts are utilized in modeling the user behavior. The methodology is presented via a use case scenario. Data collected from various ubiquitous sensors are used in recognizing and defining contexts, and association rules determine the routines. The focus is on testing the suitability of the apriori algorithm for this application area. Several useful routines were derived from the user data, and the results show that data mining can be utilized in pervasive computing.
Keywords :
data mining; knowledge based systems; learning (artificial intelligence); mobile handsets; ubiquitous computing; apriori algorithm; association rules; data mining; machine learning; pervasive computing; routine learning; temporal context sequence; ubiquitous sensors; user behavior modeling; Algorithm design and analysis; Association rules; Calendars; Context modeling; Data analysis; Data mining; Intelligent systems; Machine learning algorithms; Pervasive computing; Testing;
Conference_Titel :
Information Technology: Coding and Computing, 2004. Proceedings. ITCC 2004. International Conference on
Print_ISBN :
0-7695-2108-8
DOI :
10.1109/ITCC.2004.1286633
