Capturing Regular Human Activity through a Learning Context Memory
Autor: | Mohr, P. H., Ryan, N., Jonathan Ian Timmis |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2006 |
Předmět: | |
Zdroj: | 3rd International Workshop on Modeling and Retrieval of Context Scopus-Elsevier |
Popis: | A learning context memory consisting of two main parts is\ud presented. The first part performs lossy data compression,\ud keeping the amount of stored data at a minimum by combining\ud similar context attributes — the compression rate for the\ud presented GPS data is 150:1 on average. The resulting data is\ud stored in an appropriate data structure highlighting the level\ud of compression. Elements with a high level of compression\ud are used in the second part to form the start and end points\ud of episodes capturing common activity consisting of consecutive\ud events. The context memory is used to investigate how\ud little context data can be stored containing still enough information\ud to capture regular human activity. |
Databáze: | OpenAIRE |
Externí odkaz: |