Semantic Annotation of Global Positioning System Traces Activity Type Inference

Abstract

Because of the rapid development of technology, larger data sets on activity travel behavior have become available. These data sets often lack semantic interpretation. This lack of interpretation implies that annotation of activity type and transportation mode is necessary. This paper aims to infer activity types from Global Positioning System (GPS) traces by developing a decision tree-based model. The model considers only activity start times and activity durations. On the basis of the decision tree classification, a probability distribution and a point prediction model were constructed. The probability matrix described the probability of each activity type for each class (i.e., combination of activity start time and activity duration). In each class, the point prediction model selected the activity type that had the highest probability. Two types of data were collected in 2006 and 2007 in Flanders, Belgium (i.e., activity travel data and GPS data). The optimal classification tree constructed contained 18 leaves. Consequently, 18 if-then rules were derived. An accuracy of 74% was achieved when the tree was trained. The accuracy of the model for the validation set (72.5%) showed that overfitting was minimal. When the model was applied to the test set, the accuracy was almost 76%. The models indicated the importance of time information in the semantic enrichment process. This study contributes to future data collection in that it enables researchers to infer activity types directly from activity start time and duration information obtained from GPS data. Because no location information is needed, this research can be easily and readily applied to millions of individual agents.