An Exemplary Survey Implementation on Text Mining with Rapid Miner

Abstract

Recently many disciplines such as databases, statistics, and information retrieval have affected the growth of data mining. By this phenomenon, data mininghad been allowing to extract information or knowledge from these data. Data mining employs diverse techniques based on this phenomenon, but most existing approach is data analysis and text mining. Text Mining allows various analyses for data representing the new framework known as Information Extraction. The extraction ofinformation from unstructured resources has released new paths for analyzing,organizing and querying data. Adapting and implementing these patterns has beentime-consuming in the past but with the use of some discovery tools now this taskhas been significantly easier. The first part of the study includes the mining of an unstructured text and the second part is visualization, both provided by Rapid Miner. The visualization part has some subparts which are directly related to the conducted survey dedicated to particular individuals. These subparts are; finding correlation between the attributes of people, determination of the most weighted attribute and clustering subpart which classifies a group of people that has moresimilarity between them. Each of these parts has been examined and shown by definitions, examples, analysis and conclusions.