Mining Comprehensible and Interesting Rules: A Genetic Algorithm Approach

Abstract

A majority of contribution in the domain of rule mining overemphasize on maximizing the predictive accuracy of the discovered patterns. The user-oriented criteria such as comprehensibility and interestingness are have been given secondary importance. Recently, it has been widely acknowledged that even highly accurate discovered knowledge might be worthless if it scores low on the qualitative parameters of comprehensibility and interestingness. This paper presents a classification algorithm based on evolutionary approach that discovers comprehensible and interesting in CNF form in which along with conjunction in between various attributes there is disjunction among the values of an attribute. A flexible encoding scheme, genetic operators with appropriate syntactic constraints and a suitable fitness function to measure the goodness of rules are proposed for effective evolution of rule sets. The proposed genetic algorithm is validated on several datasets of UCI data set repository and experimental results are presented which clearly indicate lower error rates and more comprehensibility across a range of datasets. Some of the rules show the interesting and valuable nuggets of knowledge discovered from small disjuncts of high accuracy and low support which are very difficult to capture otherwise.

Knowledge Acquisition tool for Classification Rules using Genetic Algorithm Approach

“…Mutation operation is used to guarantee that the search is done in the whole population by maintaining the diversity of gene. Higher mutation probability can turn the genetic algorithm in a random search algorithm [14]. In the present work, the mutation probability (Pm) = 0.01.…”

“…The CRM model with GA in [11] considers the characteristics of cloud computing. In [14] a classification algorithm based on GA approach presented to discover production rules in Conjunctive Normal Form (CNF) where a conjunctive relationship exists between two attributes and disjunction is there among the values of the same attribute. In [17,20] the proposed GA for classification IF-THEN rules tried to avoid the drawbacks of creating randomly an initial population by creating initial population in a systematic way using the generalized Uniform Population (UP) method.…”

“…Selection, crossover and mutation are the basic genetic operators for generating offspring from the fittest individuals. Several GA approaches have been designed for discovering classification rules [11][12][13][14][15][16][17][18][19]. The contribution of this paper is the discovery of significant (novel) rules from large dataset using genetic algorithm approach.…”

Probabilistic logic reasoning for subjective interestingness analysis

“…Classi cation rule mining aims at discovering a set of implication patterns that relates certain object features (attributes) to a label representing the category of the object under analysis (Vashishtha et al;2011). Let X = {X 1 ..., Xn} denote a set of variables whose elements identify the attributes 1 used to describe the objects to be classi ed.…”

Particle Swarm Optimization based Feature Selection

“…Classification is an important task of data mining in which a model is constructed from training set and this model is used to classify unseen data. Typically, a real world dataset comprising of irrelevant and redundant features create problem in classification and make classification a really challenging task [1], [2]. Feature Selection is capable of tackling this challenge by selecting relevant features and improving the accuracy of classification and other data mining tasks.…”