Vol. 2, Issue 5, Part I (2016)

Abstract
In the real world, data are represented in the form of facts, numbers and text. The data are accumulated in vast quantity and have also grown in different formats and databases. They are categorized as operational, transactional, nonoperational and metadata. These data are analyzed and the knowledge is derived from the original data using data mining. Data Mining is a process of analyzing the data in different perspectives such as association, clustering, classification and regression, prediction. Web data mining can be defined as the discovery and analysis of data from all over the world. World Wide Web has huge volume of data, which may be very useful or sometimes may be useless Meta data. Web analytics is the measurement, collection, analysis and reporting of internet data for purposes of understanding and optimizing web usage. The information is collected from the analysis in the form of patterns, association or relationships among the data. The collected information is converted into knowledge, which is gathered from the historical terms and is applied as future trends. Data can be collected from the different repositories. They may contain noisy data, redundant, irrelevant and insignificant features. In this scenario, web sentiment prediction plays a vital role of identifying the relevant predictions and data from the dataset. In this article deals concept of data mining, web mining, Information Retrieval, web log mining and opinion mining.