Development Of Decision Related Engine Using Integration Of Genetic Algorithm And Text Mining

Abstract

Text Miningas a popular method used to perform text-based information retrieval can not be denied its use. At the beginning of the text mining method is based on the purpose of seeking to represent a collection of words from a document. Tokenizing, filtering, Stemming, Tagging and Analyzing are the critical process sequence in the methods of Text Mining, which ultimately yield important information of the most dominant in a document  that analyzed.

Applying Genetic Algorithms on Text Mining methodaims to create a Related Decision Engine, a machine capable of making decisions that relate among one another and the results of analysis carried out. Analysis carried out in this study is the analysis applied to digital libraries, the information starting from the synopsis, preface and the title of the book, which will be processed the information by the methods of Text Mining. Users often only presented the information in accordance with what books you are looking for, without giving a few other books that likely her relationship with the books to be searched. Though likely will also be of interest to the user associated with the book that will be sought. In this case, the users who will find the documents were also given another information that her relationship with books that may also be an option in the user's search.

This application created a web-based, using the Apache Web Server and MySQL database to support the completeness of applications built on digital libraries.