Analysis Of Data Mining Methods Naive Bayes Classifier (NBC)

Abstract

At present, the development of the information that is unbelievably experienced rapid development is no exception to the application database. In a database application is needed to store important data and the development of increasingly large storage media so that data can be stored on a database of many in the data storage media. Indirectly lead to the accumulation of data and unwittingly data it stores a lot of information that can be utilized. Therefore, the necessary data mining, which is a concept to discover knowledge hidden within a database. Data mining is a semi-automatic process that uses statistical techniques, mathematics, artificial intelligence, and machine learning to extract and identify potential knowledge and useful information that is stored in large databases. Find information hidden within a large data one method that can be used is a method of Naive Bayes classifier (NBC). Methods Naive Bayes classifier (NBC) is one method that is likely a simple data classification based on Bayes theorem, assuming between the explanatory variables are independent (independent). This method is one of many methods used to classification hidden data and get information from a lot of data. Naive Bayes classifier method assumes that the presence or absence of a particular occurrence of a group is not associated with the presence or absence of other events. Naive Bayes classifier method can be combined with a Decision Support System for the method does not require weights to do the calculations, but only using the probability data that already exists. 2 Reasons for choosing Naive Bayes classifier method in this study because of the classification of the opportunities that simple and easy to understand, efficient computation, and the results can be recognized more accurately than other methods, and has the ability to classification by category in a simple mathematical form. Thus it is expected that this method will be able to find information hidden from much of the data with the classification by category