Exploring the Usefulness of a Decision Tree in Predicting People's Locations

Location prediction has been investigated by many researchers. However, there are few studies that explore the relationship between human activities and location. This paper proposes the use of a decision tree to investigate how location can be predicted after observing a series of human activities. A decision tree provides a set of easy-to-interpret decision rules that are necessary for decision makers to be able to make timely and appropriate decisions about location prediction. Based on more than 6000 contextual datasets obtained from college students, we conducted experiments with the WEKA software. Our findings revealed that, given a number of human activities and personal information, a decision tree classifier provides a set of useful rules through which appropriate inferences about location prediction are possible.