Abstract
The focus of this paper is to learn the daily activity engagement patterns of travelers by using a nontraditional model called Support Vector Machines (SVM) that is widely used in Artificial intelligence and Machine Learning. It is postulated that an individualâ??s choice of activities depends not only on socio-demographic characteristics but also on previous activities of individual at the same day. In the paper, Markov Chain models are used to study the sequential choice of activities. The dependency between activity type, activity sequence and socio-demographic data are captured by employing Conditional Random Fields. In order to learn model parameters, we use sequential multinomial logit model and multiclass Support Vector Machines (K-SVM) with two different dependency structures. In the first dependency structure, it is assumed that type of activity at time t depends on the last previous activity and sociodemographic data, whereas in the second structure we assume activity selection at time t depends on all previous activity types of the individual on the same day and her sociodemographic characteristics. The models are applied to data drawn from Orange County and San Diego County households and a comparison of the accuracy of estimation indicates the superiority of K-SVM with first dependency structure over the other models tested. Additionally, we show that by using different sets of explanatory variables or tuning parameters of the kernel function in K-SVM, its accuracy in estimating activity patterns increases.