期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2013
卷号：2
期号：11
页码：2956-2962
出版社：Shri Pannalal Research Institute of Technolgy
摘要：In data mining many techniques are available for predicting of frequent pattern. One technique is association rule mining algorithm which can solve critical problem in biological field but this algorithm has limitation of space, time complexity and accuracy. Classification rules are one kind of conditional rules which can be used to discover data from large data sets. The model that is built is called a Classifier or Predictor depending upon whether the model finds the unknown data class or data value. Association analysis is the task of bringing out relationship among data. Association analysis is most popular analysis technique in data mining for classifying biological data. When dealing with Biological data large search spaces may arise. Genetic algorithms deal with large search space very effectively. Combining association rule mining and Genetic algorithms to classify biological data is a novel and extensive research area. The purpose of this thesis is to classify biological data with association rule and genetic algorithm. A Genetic Algorithm (GA) is an iterative search, optimization and adaptive machine learning technique premised on the principles of Natural Selection. A GA is a search method that functions analogously to an evolutionary process in a biological system as it mimics evolution and competition between individuals in natural selection. It generates a better solution from existing solutions. Neither programmer nor genetic algorithm has to know how to solve a given problem; solution is just bred. GAs is one of the most robust problem solving techniques. They can find solutions of NP-hard problems easily. For problems with a larger parameter space and where the problem itself can be easily specified, GA can be an appropriate solution. "Genetic Algorithms are software procedures modeled after genetics and evolution".
