Extracting Places and Activities from GPS Traces Using Hierarchical Conditional Random Fields

摘要: Learning patterns of human behavior from sensor data is extremely important for high-level activity inference. This paper describes how to extract a person's activities and significant places traces GPS data. The system uses hierarchically structured conditional random fields generate consistent model places. In contrast existing techniques, this approach takes the context into account in order detect person. Experiments show improvements over techniques. Furthermore, they indicate that proposed able robustly estimate using trained collected by other persons.