The idea of using composite features has been well studied in the field of machine learning and goes under the name of
constructive induction. Constructive induction is a process of creating new features/attributes from the task-supplied
attributes and then building a model on both these new as well as task supplied attributes [S.78]. For most cases
this approach is diametrically opposite of dimensionality reduction; dimensionality reduction tries to eliminate attributes/
features whereas constructive induction expands the feature space before building the classification model.
There are many ways of creating new features, Zheng et al [Zij96] presents a discussion of using conjunctive, disjunctive
and x of N features. The features of type x of N were first studied by Murphy et al [MP91]. Brodley et al [BU92]
consider composite features which are modeled as linear functions, which operate on different attribute values. In
this paper we will be limiting ourselves to the study of conjunctive attribute values, a detailed discussion about the
advantages of using conjunctive attributes in context of different classifiers is presented in Section 5.