Automatic Induction of Classification Rules from Examples Using N-Prism

Abstract

One of the key technologies of data mining is the automatic induction of rules from examples, particularly the induction of classification rules. Most work in this field has concentrated on the generation of such rules in the intermediate form of decision trees. An alternative approach is to generate modular classification rules directly from the examples. This paper seeks to establish a revised form of the rule generation algorithm Prism as a credible candidate for use in the automatic induction of classification rules from examples in practical domains where noise may be present and where predicting the classification for previously unseen instances is the primary focus of attention.

A Scalable Expressive Ensemble Learning Using Random Prism: A MapReduce Approach

“…'Divide and conquer' based classifiers produce a decision tree, such as Quinlan's C4.5 decision tree induction algorithm [18]; 'separate and conquer' based classifiers produce a set of IF...THEN classification rules, such as the Prism family of algorithms [8,4,5]. As pointed out in [20], most ensemble classifiers are based on the 'divide and conquer' approach even though Prism classifiers have shown to be less vulnerable to overfitting compared with decision tree based classifiers [4]. This is especially the case when confronted with noise and missing values in the data [4].…”

“…As pointed out in [20], most ensemble classifiers are based on the 'divide and conquer' approach even though Prism classifiers have shown to be less vulnerable to overfitting compared with decision tree based classifiers [4]. This is especially the case when confronted with noise and missing values in the data [4]. A recently developed ensemble classifier named Random Prism [20], which is inspired by RF and RDF, makes use of the Prism family of algorithms as base classifiers.…”

Random Prism: An Alternative to Random Forests

“…'Separate and conquer' can be traced back to Michalski's AQ system in the 1960s [16]. However the most notable development using the 'separate and conquer' approach is the Prism family of algorithms [8,3,4]. It produces modular rules that do not necessarily fit into a decision tree.…”

“…Recent developments on the Prism family of algorithms includes frameworks for parallelising Prism algorithms for rule induction on massive datasets [23,25,24] and rule pruning methods in order to reduce overfitting [28,4]. In general Prism algorithms have been shown to be less vulnerable to overfitting compared with decision tree classifiers, especially if there is noise in the data and missing values [3]. Yet most ensemble learning approaches are either based on decision trees or a heterogeneous setup of base classifiers.…”

P-Prism: A Computationally Efficient Approach to Scaling up Classification Rule Induction

“…The aim is to generate rules with significantly fewer redundant terms than those derived from decision trees. Compared with decision trees Prism [1]:…”