Investigate the Impact of Dataset Size on the Performance of Data Mining Algorithms

Bondu Venkateswarlu; Dr.GSV Prasad Raju

Abstract

KEEL (Knowledge Extraction based on Evolutionary Learning) tool is used to analyze the datasets to access the performance of various existing data mining algorithms. KEEL is an open source tool that can be used for a large no of knowledge data discovery task. It provides a simple GUI based and data flow to design experiments with different datasets to access the behavior of the various existing data mining algorithms. With the combination of classification Algorithms and Bayesian-D preprocessing technique used in KEEL tool, we analyze the performance of classification algorithms by varying the size of dataset records from 500 to 5000. We investigate the impact of dataset size on global classification error, standard deviation global classification error and correctly classified for both training and testing for the classification algorithms such as C4.5-C, AdaBoost-C and C4.5_Binirization-C. From the experimental result reveals the C4.5-C out performed and also found that by varying size from 500 to 5000 the variance of global classification error is 0.001727, standard deviation global classification error is 0.004158 and correctly classified is 0.998267.

Keywords

KEEL, C4.5-C, Adaboost, C4.5_BinirizationC,GUI

References

[1]. “Novel Centroid Selection Approaches For KmeansClustering Based Recommender Systems”, Sobia Zahra a, Mustansar Ali Ghazanfar a, Asra Khalid a, Muhammad Awais Azam a,Usman Naeem b, Adam Prugel-Bennett c, doi 10.1016 2015.03.062, 0020-0255/ 2015 Elsevier .

[2]. “Analysis and Classification of Hardwood Species based on Coiflet DWT Feature Extraction and WEKA Workbench”, Arvind R. Yadav1, R. S. Anand 2, M. L. Dewal3, Sangeeta Gupta4 978-1-4799-2866-8/14/ 2014 IEEE

[3]. “Challenges In Knowledge Discovery And Data Mining In Datasets”, Mykhaylo Lobur1, Yuri Stekh2, Vitalij Artsibasov3 IEEE May 2011

[4]. “Spectral clustering and semi-supervised learning using evolving similarity graphs”, Christina Chrysouli, Anastasios Tefas, doi/10.1016/j. asoc.2015.05.0261568-4946/ 2015 Elsevier

[5]. “Exploration of Soft Computing Models for the Valuation of Residential Premises using the KEEL Tool”, Tadeusz Lasota, Ewa Pronobis, Bogdan TrawiÅ„ski, Krzysztof TrawiÅ„ski, 978-0-7695-3580-7/09 2009 IEEE.

[6]. “KEEL: A data mining software tool integrating genetic fuzzy systems”, Jesus Alcala-Fdez, Salvador GarcÄ±a, Francisco Jose Berlanga , Alberto Fernandez, Luciano S ´ anchez, M.J. del Jesus and Francisco Herrera, 978-1-4244-1613-4/08/2008IEEE

[7]. “KEEL: a software tool to assess evolutionary algorithms for data mining problems”, J. Alcalá-Fdez · L. Sánchez · S. García · M. J. del Jesus ·,S. Ventura · J. M. Garrell · J. Otero · C. Romero · J. Bacardit · V. M. Rivas · J. C. Fernández · F. Herrera Published online: 22 May 2008 © Springer-Verlag 2008

[8]. “Investigation of Fuzzy Models for the Valuation of Residential Premises using the KEEL Tool”, Tadeusz Lasota, Jacek Mazurkiewicz, Bogdan TrawiÅ„ski, and Krzysztof TrawiÅ„ski, 978-0-7695-3326-1/08 2008 IEEE.

[9]. “Implementation and Integration of Algorithms into the KEEL Data-Mining Software Tool”, Alberto Fernandez, Juli´ an Luengo, Joaquin Derrac, Jes´ us Alcal´ a-Fdez, and Francisco Herrera H. Yin and E. Corchado (Eds.): IDEAL 2009, LNCS 5788, pp. 562–569, 2009. c Springer-Verlag Berlin Heidelberg 2009

[10]. “Comparison of Various Classification Techniques Using Different Data Mining Tools for Diabetes Diagnosis”, Rashedur M. Rahman, Farhana Afroz Journal of Software engineering and Applications, 2013, 6, 85-97 doi/10.4236/jsea.2013.63013 Published Online March 2013

[11]. “Data Mining Techniques and Their Implementation in Blood Bank Sector –A Review”, Ankit Bhardwaj, Arvind Sharma, V.K. Shrivastava / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 Vol. 2, Issue4, JulyAugust 2012, pp.1303-1309 1303.

[12]. “Application of Knowledge Discovery in Database to Blood Cell Counter Data to Improve Quality Control in Clinical Pathology”, D.Minnie, S.Srinivasan 978-0-7695-4514-1/11 IEEE.

[13]. “Classifying Blood Donors Using Data Mining Techniques” P.Ramachandran, Dr.N.Girija, 3Dr.T.Bhuvaneswari, IJCSET Feb 2011 Vol 1, Issue 1,10-13

[14]. “Comparison of Various Classification Techniques Using Different Data Mining Tools for Diabetes Diagnosis”, Rashedur M. Rahman, Farhana Afroz Journal of Software engineering and Applications, 2013, 6, 85-97 doi/10.4236/jsea.2013.63013 Published Online March 2013

[15]. “Data Mining to Improve Safety of Blood Donation Process “, Madhav Erraguntla , Peter Tomasulo, Kevin Land, Hany Kamel, Marjorie Bravo, Barbee Whitaker, Richard Mayer, Sarita Khaire 978-1-4799-2504-9/14 2014 IEEE

Cites this article as

B. Venkateswarlu, D. P. Raju, "Investigate the Impact of Dataset Size on the Performance of Data Mining Algorithms", International Journal of Innovative Research in Engineering & Management (IJIREM), Vol-3, Issue-2, Page No-91-98, 2016. Available from: