Discovery of Classification Rules Using Distributed Genetic Algorithm

Classification is an important topic in data mining research. Given a set of data records, each of which belongs to one of a number of predefined classes, the classification problem is concerned with the discovery of classification rules that can allow records with unknown class membership to be correctly classified. Many algorithms have been developed to mine large data sets for classification models and they have been shown to be very effective. However, when it comes to determining the likelihood of each classification made, many of them are not designed with such purpose in mind. For this, they are not readily applicable to such problems as churn prediction. For such an application, the goal is not only to predict whether or not a subscriber would switch from one carrier to another, it is also important that the likelihood of the subscriber's doing so be predicted. The reason for this is that a carrier can then choose to provide a special personalized offer and services to those subscribers who are predicted with higher likelihood to churn. Given its importance, we propose a new data mining algorithm, called data mining by evolutionary learning (DMEL), to handle classification problems of which the accuracy of each predictions made has to be estimated. In performing its tasks, DMEL searches through the possible rule space using an evolutionary approach that has the following characteristics: 1) the evolutionary process begins with the generation of an initial set of first-order rules (i.e., rules with one conjunct/condition) using a probabilistic induction technique and based on these rules, rules of higher order (two or more conjuncts) are obtained iteratively; 2) when identifying interesting rules, an objective interestingness measure is used; 3) the fitness of a chromosome is defined in terms of the probability that the attribute values of a record can be correctly determined using the rules it encodes; and 4) the likelihood of predictions (or classifications) made are estimated so that subscribers can be ranked according to their likelihood to churn. Experiments with different data sets showed that DMEL is able to effectively discover interesting classification rules. In particular, it is able to predict churn accurately under different churn rates when applied to real telecom subscriber data.

This paper deals with the simplification of classification rules for data mining using rough set theory combined with the hierarchical granulation structure. In the proposed method, the procedure for a classification rule discovery from data consists of two parts; the reduction of attributes and the rule discovery. Rough set theory is used to classify the objects of interest into the similarity classes and to investigate the granularity of knowledge for reasoning of uncertain concepts, and the hierarchical granulation structure is adopted to find the classification rules effectively. The proposed classification method generates minimal classification rules and an explicit and effective structure is achieved in consequence. Also the computational burden for the classification rule discovery is considerably reduced. Therefore it may offer an easy way to analyze the information system. To show the effectiveness of the proposed method, a simulation is performed on Wisconsin Breast Cancer data. The simulation result shows that the proposed method gives a good performance in spite of very simple rules and short conditionals.

Evolutionary Algorithms (EAs) are stochastic search algorithms inspired by the process of neo-Darwinian evolution. The motivation for applying EAs to data mining is that they are robust, adaptive search techniques that perform a global search in the solution space. This chapter first presents a brief overview of EAs, focusing mainly on two kinds of EAs, viz. Genetic Algorithms (GAs) and Genetic Programming (GP). Then the chapter reviews the main concepts and principles used by EAs designed for solving several data mining tasks, namely: discovery of classification rules, clustering, attribute selection and attribute construction. Finally, it discusses Multi-Objective EAs, based on the concept of Pareto dominance, and their use in several data mining tasks.

Data mining aims to discover interesting, non-trivial, and meaningful information from large datasets. One of the data mining tasks is classification, which aims to assign the given datasets to the most suitable classes. Classification rules are used in many domains such as medical sciences, banking, and meteorology. However, discovering classification rules is challenging due to large size and noisy structure of the datasets, and the difficulty of discovering general and meaningful rules. In the literature, there are several classical and heuristic algorithms proposed to mine classification rules out of large datasets. In this paper, a new and novel heuristic classification data mining approach based on artificial bee colony algorithm (ABC) was proposed (ABC-Miner). The proposed approach was compared with Particle Swarm Optimization (PSO) rule classification algorithm and C4.5 algorithm using benchmark datasets. The experimental results show the efficiency of the proposed method.

Existing music recommendation systems don't suit user subjectivity and are not naturally intuitive. We previously proposed a music recommendation system to address these problems. Our music recommendation system uses Kansei agents and a Kansei retrieval system. The Kansei agents employ a Kansei model and are characterized by a three-layered neural network. Music fluctuation properties are defined as inputs to the neural networks of the Kansei agents. Our previously proposed system returns recommendations based on user subjectivity. However, this system incurs significant user evaluation load. In this study, we introduce a distributed genetic algorithm (DGA) as an optimization method to address this problem. We compare the proposed DGA method to our previous method by simulation. The simulation results indicate that the proposed DGA method significantly outperforms previous method.

Previous studies show that using significant classification rules to accomplish the classification task is suitable for bio-medical research. Discovery of many significant rules could be performed by using ensemble methods in decision tree induction. However, those traditional approaches are not useful for incremental task. In this paper, we use an ensemble method named Cascading and Sharing to derive many significant classification rules from incrementally inducted decision tree and improve the classifiers accuracy.

Integrating Internet technologies with traditional healthcare systems has enabled the emergence of cloud healthcare systems. These systems aim to optimize the balance between online diagnosis and offline treatment to effectively reduce patients’ waiting times and improve the utilization of idle medical resources. In this paper, a distributed genetic algorithm (DGA) is proposed as a means to optimize the balance of patient assignment (PA) in cloud healthcare systems. The proposed DGA utilizes individuals as solutions for the PA optimization problem and generates better solutions through the execution of crossover, mutation, and selection operators. Besides, the distributed framework in the DGA is proposed to improve its population diversity and scalability. Experimental results demonstrate the effectiveness of the proposed DGA in optimizing the PA problem within the cloud healthcare systems.

Privacy preserving rule-based classifier using modified artificial bee colony algorithm

In this paper, we evaluate the performance of a distributed genetic algorithm (DGA) with cellular structures on function optimization problems. The DGA in this paper is a genetic algorithm with several subpopulations. Each subpopulation in the DGA has a cellular structure as in cellular genetic algorithms and has interaction area with other subpopulations. We examine several parameter specifications in the DGA on well-known function optimization problems.

In this paper, we consider a two-level 0-1 programming problem in which there is not coordination between the decision maker (DM) at the upper level and the decision maker at the lower level. We propose a revised computational method that solves problems related to computational methods for obtaining the Stackelberg solution. Specifically, in order to improve the computational accuracy of approximate Stakelberg solutions and shorten the computational time of a computational method implementing a genetic algorithm (GA) proposed by the authors, a distributed genetic algorithm is introduced with respect to the upper level GA, which handles decision variables for the upper level DM. Parallelization of the lower level GA is also performed along with parallelization of the upper level GA. The proposed algorithm is also improved in order to eliminate unnecessary computation during operation of the lower level GA, which handles decision variables for the lower level DM. In order to verify the effectiveness of the proposed method, we propose comparisons with existing methods by performing numerical experiments to verify both the accuracy of the solution and the time required for the computation.

In data mining, nugget discovery is the discovery of interesting classification rules that apply to a target class. In previous research, heuristic methods (genetic algorithms, simulated annealing and tabu search) have been used to optimise a single measure of interest. This paper proposes the use of multi-objective optimisation evolutionary algorithms to allow the user to interactively select a number of interest measures and deliver the best nuggets (an approximation to the Pareto-optimal set) according to those measures. Initial experiments are conducted on a number of databases, using an implementation of the fast elitist non-dominated sorting genetic algorithm (NSGA), and two well known measures of interest. Comparisons with the results obtained using modern heuristic methods are presented. Results indicate the potential of multi-objective evolutionary algorithms for the task of nugget discovery.

A major problem in the use of genetic algorithms is premature convergence. One approach for dealing with this problem is the distributed genetic algorithm model. Its basic idea is to keep, in parallel, several subpopulations that are processed by genetic algorithms, with each one being independent of the others. Making distinctions between the subpopulations by applying genetic algorithms with different configurations, we obtain the so-railed heterogeneous distributed genetic algorithms. These algorithms represent a promising way for introducing a correct exploration/exploitation balance in order to avoid premature convergence and reach approximate final solutions. This paper presents the gradual distributed real-coded genetic algorithms, a type of heterogeneous distributed real-coded genetic algorithms that apply a different crossover operator to each sub-population. Experimental results show that the proposals consistently outperform sequential real-coded genetic algorithms.

Achieving good performance with a parallel genetic algorithm requires properly configuring control parameters such as mutation rate, crossover rate, and population size. We consider the problem of setting control parameter values in a standard, island-model distributed genetic algorithm. As an alternative to tuning parameters by hand or using a self-adaptive approach, we propose a very simple strategy which statically assigns random control parameter values to each processor. Experiments on benchmark problems show that this simple approach can yield results which are competitive with homogeneous distributed genetic algorithm using parameters tuned specifically for each of the benchmarks.

We study classification over a slow stream of complex objects like customers or students. The learning task must take into account that an object’s label is influenced by incoming data from adjoint, fast streams of transactions, e.g. customer purchases or student exams, and that this label may even change over time. This task involves combining the streams, and exploiting associations between the target label and attribute values in the fast streams. We propose a method for the discovery of classification rules over such a confederation of streams, and we use it to enhance a decision tree classifier. We show that the new approach has competitive predictive power while building much smaller decision trees than the original classifier.KeywordsRule MiningFrequent ItemsetsAssociation Rule MiningHigh Predictive PowerSlow StreamThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The paper presents a distributed genetic algorithm implementation for obtaining good quality consistent results for different ordering problems. Most importantly, the solution found by the proposed Distributed GA is not only of high quality but also robust and does not require fine tuning of the probabilities of crossover and mutation. In addition, implementation of the Distributed GA is simple and does not require the use of any specialized, expensive hardware. Fault tolerance has also been provided by dynamic reconfiguration of the distributed system in the event of a process or machine failure. The effectiveness of using a simple crossover scheme with Distributed GA is demonstrated by solving three variations of the Traveling Salesman Problem (TSP).