Prototype survey analysis of different information retrieval classification and grouping approaches for categorical information

Abstract

Present days, large amount information stored in information sources, which is formally increased based on Knowledge Discovery from information warehouses with different formats of information. To acquire required and useful information from information sources, some of the techniques, methods and some of developed tools to combine large or high dimensional information sets. This procedure gives demand to implement novel research field i.e. information retrieval. The main task behind information retrieval is to extract required information from large size of information and change them into meaningful for further use in information retrieval. Classification and Grouping's are the main information retrieval approaches to classify and combine categorical information in a large set of information into required group set of class labels. So in this document, we provide comprehensive analysis of different classification and grouping methods in information retrieval to efficient information retrieval, which includes neural networks, Bayesian networks and decision trees. We also provide survey on some of semi supervised and supervised outlier detection techniques for categorical information on unlabeled information sets under large of instances in information sets with required instances in real time synthetic information. We bring out the keys aspects of different outlier and information retrieval approaches to information exploration.