Classification of High Dimensional Data Set Using Rough Set Theory RST and Support Vector Machine SVM

Abstract

Artificial intelligence is becoming a need of day to day life. In the domain of essentials, all the things are required to be artificial intelligent. To meet the requirement one important step is to find an intelligent way of computer and user interaction. Data collection and classification both are part of artificial intelligence. newlineMachine learning is considered as a subset of artificial intelligence. In the Machine learning process a computer learns from the given dataset. After learning from the dataset it predicts the results of new data. The new data is similar to the training data. Machine learning process is all about performing special tasks by generalizing from the examples. In machine learning the focuses is given to development of computer programs which are capable of teaching itself to grow and changing when exposed to the new data. These systems are capable to automatically learn from the given data. newlineMachine learning algorithm identifies patterns based on different features and then makes predictions on new unclassified data based on the patterns which it learned earlier. It is basically an algorithm that approximates a system response with a learned function. Developing some general purpose algorithm of practical value is the main goal of machine learning research. It can be considered as learning by examples which mean that it can find out how to perform important tasks by generalizing from examples. As more and more data becomes available more complex problems can be easily solved. Machine learning is about learning to do better in the future base on past experiences. Supervised and unsupervised learning methods are two approaches of Machine learning methods. newlineFor different applications Classification is becoming one of the most important tasks. Classification is one of the most important tasks for different applications. Most of the existing supervised classification methods are based on traditional statistics, which can provide ideal results when sample size is tending to infinity. However, only fi