Parametric Action Decision Trees: Incorporating Continuous Attribute Variables Into Rule-Based Models of Discrete Choice

Rule-based models, such as decision trees, are ideally suited to represent discontinuous effects of independent variables on discrete choice behavior in transport or spatial systems. At the same time, however, the models require that continuous attributes, such as for example travel time and travel costs, are discretisized, which may decrease the sensitivity of predictions for policy measures that involve these attributes. To overcome this problem and combine the specific strengths of the rule-based and parametric modeling approaches, this paper introduces a hybrid approach. The so-called parametric action decision tree (PADT) replaces the conventional action-assignment rule of the decision tree by a logit model or any other parametric discrete choice model. The PADT includes alternative-specific constants to take the impact of leaf-node membership into account in addition to terms for the continuous attributes. As an illustration, we show how the approach can be used to incorporate travel-costs sensitivity in Albatross - a rule-based model of activity-travel choice. The results indicate that the enhanced, hybrid model can reproduce realistic ranges of price elasticities of travel demand.

[1]  Geert Wets,et al.  Association Rules in Identification of Spatial-Temporal Patterns in Multiday Activity Diary Data , 2001 .

[2]  Toshiyuki Yamamoto,et al.  Drivers’ Route Choice Behavior: Analysis by Data Mining Algorithms , 2002 .

[3]  Jean-Claude Thill,et al.  Tree Induction of Spatial Choice Behavior , 1999 .

[4]  Ta Theo Arentze,et al.  Reinduction of Albatross Decision Rules with Pooled Activity-Travel Diary Data and an Extended Set of Land Use and Cost-Related Condition States , 2003 .

[5]  Chi Xie,et al.  WORK TRAVEL MODE CHOICE MODELING USING DATA MINING: DECISION TREES AND NEURAL NETWORKS , 2002 .

[6]  Larry Lucardie,et al.  DECISION-MAKING UNDER NOT-WELL-DEFINED CONDITIONS: FROM DATA PROCESSING TO LOGICAL MODELLING , 1991 .

[7]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[8]  Joffre Swait,et al.  A NON-COMPENSATORY CHOICE MODEL INCORPORATING ATTRIBUTE CUTOFFS , 2001 .

[9]  D. Graham,et al.  The Demand for Automobile Fuel A Survey of Elasticities , 2002 .

[10]  Tommy Gärling,et al.  Computational-Process Modelling of Household Activity Scheduling , 1993 .

[11]  Harry J. P. Timmermans,et al.  Measuring the goodness-of-fit of decision-tree models of discrete and continuous activity-travel choice: methods and empirical illustration , 2003, J. Geogr. Syst..

[12]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[13]  David A. Landgrebe,et al.  A survey of decision tree classifier methodology , 1991, IEEE Trans. Syst. Man Cybern..

[14]  Harry Timmermans,et al.  A hybrid compensatory-noncompensatory model of residential preference structures , 1986 .

[15]  Geert Wets,et al.  Identifying Decision Structures Underlying Activity Patterns: An Exploration of Data Mining Algorithms , 2000 .

[16]  A. Tversky.  Elimination by aspects: A theory of choice. , 1972 .

[17]  Davy Janssens,et al.  Improving Performance of Multiagent Rule-Based Model for Activity Pattern Decisions with Bayesian Networks , 2004 .

[18]  M. Shaw,et al.  Induction of fuzzy decision trees , 1995 .

[19]  Davy Janssens,et al.  Integrating Bayesian networks and decision trees in a sequential rule-based transportation model , 2006, Eur. J. Oper. Res..

[20]  Theo Arentze,et al.  Measuring Impacts of Condition Variables in Rule-Based Models of Space-Time Choice Behavior: Method and Empirical Illustration , 2002 .

[21]  Ta Theo Arentze,et al.  Experiences with developing ALBATROSS: a learning-based transportation oriented simulation system , 1998 .

[22]  T. Arentze,et al.  The Impact of Simplification in a Sequential Rule-Based Model of Activity-Scheduling Behavior , 2005 .

[23]  Allen Newell,et al.  Human Problem Solving. , 1973 .

[24]  Hjp Harry Timmermans,et al.  A learning-based transportation oriented simulation system , 2004 .

[25]  J. Ortúzar,et al.  A semi-compensatory discrete choice model with explicit attribute thresholds of perception , 2005 .

[26]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .