Abstract: The paper aims at proposing the fast clustering algorithm for eliminating irrelevant and redundant data. Feature selection is applied to reduce the number of features in many applications where data has hundreds or thousands of features. Existing feature selection methods mainly focus on finding relevant features. In this paper, we show that feature relevance alone is insufficient for efficient feature selection of high- dimensional data. We define feature redundancy and propose to perform explicit redundancy analysis in feature selection. A new hypothesis is introduced that dissociate relevance analysis and redundancy analysis. A clustering based method for relevance and redundancy analysis for feature selection is developed and searching based on the selected features will be performed. While the efficiency concerns the time required to find a subset of features, the effectiveness determines the quality of the subset of features. A fast clustering-based feature selection algorithm, FAST, has been selected to be used in the proposed paper. The clustering-based strategy has a higher probability of producing a subset of useful as well as independent features. To ensure the efficiency of FAST, efficient minimum-spanning tree clustering method has been adopted. When compared with FCBF, ReliefF, with respect to the classifier, namely, the tree-based C4.5, FAST not only produces smaller subsets of features but also improves the performances by reducing the time complexity. Key terms: Clustering, Feature subset selection, Minimum Spanning Tree, T-Relevance, F- Correlation. 1. INTRODUCTION Data mining uses a variety of techniques to identify lump of information or decision-making knowledge in bodies of data, and extracting them in such a manner that they can be directly use in the areas such as decision support, estimation prediction and forecasting. The data is often huge, but as it is important to have large amount of data because low value data cannot be of direct use; it is the hidden information in the data that is useful. Data mine tools have to infer a model from the database, and in the case of supervised learning this requires the user to define one or more classes. The database contains various attributes that denote a class of tuple and these are known as predicted attributes. Whereas the remaining attributes present in the data sets are called as predicting attributes. A combination of values of these predicted attributes and predicting attributes defines a class. While learning classification rules the system has to find the rules that predict the class from the predicting attributes so initially the user has to define conditions for each class, the data mine system then constructs descriptions for the classes. Basically the system should given a case or tuple with certain known attribute values so that it is able to predict what class this case belongs to, once classes are defined the system should infer rules that govern the classification therefore the system should be able to find the description of each class [2]. Feature selection involves identifying a subset of the most useful features that produces compatible results as the original entire set of features. A feature selection algorithm is basically evaluated from the efficiency and effectiveness points of view. The time required to find a subset of features is concerned with the efficiency while the effectiveness is related to the quality of the subset of features. Some feature subset selection algorithms can effectively eliminate irrelevant features but fail to handle redundant features yet some of others can remove the irrelevant while taking care of the redundant features. A Fast clustering-based feature selection algorithm (FAST) is proposed which is based on above criterion handling redundancy and irrelevancy. [1] The Minimum Spanning tree (Kruskal’s algorithm) is constructed from the F-Correlation value which is used to find correlation between any pair of features. Kruskal's algorithm is a greedy algorithm in graph theory that finds a minimum spanning tree for a connected weighted graph. It finds a subset of the edges that forms a tree that includes every vertex, where the total weight of all the edges in the tree is minimized. 2. EXISTING SYSTEM Feature subset selection generally focused on searching relevant features while neglecting the redundant features. A good example of such feature selection is Relief, which weighs each feature according to its ability to discriminate instances under different targets based on distance-based criteria function.[9] But, Relief is ineffective in removing redundant features as the two predictive but highly correlated features are likely to be highly weighted. Relief-F [6] is an extension of the traditional Relief. This A Fast Clustering-Based Feature Subset Selection Algorithm Akshay S. Agrawal Prof. Sachin Bojewar, P.G. Scholar, Department of Computer Engg., Associate Professor, ARMIET, Sapgaon, India. VIT, Wadala. akshay1661@gmail.com