Evaluation of Rapidminer-Aplication in Data Mining Learning using PeRSIVA Model

Abstract

RapidMiner is an application that is used to analyze data quantities and qualitatively to obtain information and knowledge as expected. This software is implemented to process data using several methods or algorithms in Data Minig learning. However, when using this software, users sometimes cannot distinguish between various methods or algorithms in Data Mining. Therefore, it is necessary to evaluate to optimize the use of this software in data mining learning. This study focuses on RapidMiner evaluation of data mining learning using the Persiva model. This model consists of aspects of satisfaction, behavior, impact, and effectiveness. The data collection technique was in the form of a questionnaire with 48 subjects. Data analysis used is descriptive statistics to determine satisfaction, behavior and effects. Meanwhile, Think-Aloud Retrospective technique is used to determine the effectiveness of RapidMiner. Our findings show that users are satisfied with the results of respondents on average agreeing (80%), in the aspect of behavior and impact, the percentage results are above 80%, and the use of this application has been effective with a completion rate above 90%. So it can be concluded that by using this application in data mining learning users can easily complete tasks, and be motivated, and add insights and knowledge in relevant disciplines.