A Rule Based Approach to Modeling the Effect of Tobacco Control Policies

This paper investigates the underlying patterns in data derived from the International Tobacco Control Four Country Survey. The intent of the paper is to determine which attributes have the greatest impact on smokers' plans to quit and their attempts to quit. Rule sets are derived using decision tree models and tested on temporal data sets to assess their validity and ability to generalise. Determinant attributes that can be used to implement effective tobacco control policies are identified.