Association rules in identification of spatial-temporal patterns in multiday activity diary data

Abstract

Activity-based analysis in transportation demand forecasting is one of the most promising approaches in current transportation modeling. Travel decisions are understood as the outcome of underlying scheduling activity, resulting in large-scale interviews generating a large amount of data. Traditional techniques have been shown to be inefficient in describing the dependencies between different attributes if data sets are too large. Associations between data set attributes are described by means of association rules. The discussion outlines the description of activity-based transportation data sets through association rules for identification of spatial-temporal patterns in multiday activity diary data.