International Journal of Engineering and Advanced Technology (IJEAT) ISSN: 2249 – 8958, Volume-9 Issue-3, February 2020 3672 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: C5988029320 /2020©BEIESP DOI: 10.35940/ijeat.C5988.029320 Abstract: Around two hundred and fifty million individuals, with a major part of them being ladies influenced by diabetes. This number may ascend to 380 million by another decade. The sickness has been named as the fifth deadliest illness in the world with not a single inevitable fix to be seen. With the ascent of data innovation and proceeding with an approach into the restorative and medicinal services part, the instances of diabetes and their side effects all around are archived. Information mining is a buzz word separating concealed data from an enormous arrangement of database. It assists scientists in building large database in the area of biomedical engineering. The Pima Indian diabetes database was used for investigation purpose. In this paper an attempt has been made to study the effect of various classification and mining Techniques like Decision Tree, Naïve Bayes, SVM, Regression etc on the diagnosis of Type-2 diabetes. Keywords: Algorithms, Heart rate variability, J48, Regression, SVM I. INTRODUCTION The ECG analysis has a vital significance in analysing the various heart disorders namely, Arrhythmias, Myocardial Infarction (MI), Coronary Heart Disease (CHD), Ischemia, Cardiomyopathy, Heart attack, Aortic Aneurysm etc. The hermitian basis function and the Discrete Wavelet Transform (DWT) were some of the ECG feature extraction methods used, with which the R-R intervals were extracted for HRV analysis. The HRV analysis serves as an additional clinical and research tool for the cardiologists and researchers. In 1965 ,Professor Hon Lee has initiated the importance of HRV while doing research in the monitoring of fetal distress in women [1]. HRV finds its importance in patients with MI, Diabetes Mellitus, CHD, Chronic Heart Failure (CHF), Hypertension etc and so on. In addition to that it has been analyzed in swimmers, athletes, cyclists, foetuses and children of various age groups. Amongst all such people, HRV analysis has a marked significance in Type-2 Diabetic patients as Diabetes mellitus (DM) is major, fast growing and one of the health issues encountered by the world community. In 2030, Diabetes will be one of the leading causes of death and Type 2 Diabetes Mellitus comprises 90% of it. Hence HRV became a non-invasive tool for investigating the autonomic dysfunction related to Type-2 diabetes.[2]. Cardiology Society of European Union and the Society of Revised Manuscript Received on February 27, 2020. * Correspondence Author Sankar.P*, ECE, SV Engineering College,Karakambadi,Tirupati,AP Email: sankar.padmanaban@svcolleges.edu.in Manjunath.K.M, ECE, SV Enguneering College, Karkambadi road, Tirupathi, AP. Email: manjunath.km@svcolleges.edu.in Madhurima.V, ECE,SV Engineering college, Karkambadi road, Tirupathi, AP . Email: madhurima.v@svcolleges.edu.in Pacing and Electrophysiology in North America combine developed a task force for giving aid to these researchers. This task force developed up some standards for the measurement of Heart rate variability [3]. Based on these guidelines, this paper has been formulated by surveying and consolidating various methods, and the applications used in the HRV analysis in the current decade. The information obtained from patient records from hospitals was used to make inferences by applying several mining techniques. This mining technique helps the detection of diabetes in women at an early stage. After applying several algorithms on the database the results are obtained, compared and tabulated. The organization of the paper is given in various sections. II. HEART RATE VARIABILITY MEASUREMENTS The linear and non linear are the two methods used to analyze the Heart rate variability. the linear method always assume that the R-R interval is stationary. The R-R interval can also be mentioned as NN Interval which means that, all the processed beats are normal, without ectopic beat (skipped or extra heartbeats as compared to that of the normal ones). Due to physical or postural activities and small disturbances like Premature Ventricular Contraction heart rate can go random in nature while checking. non-linear methods are more effective than the linear methods of HRV analysis, For correct prediction in the variations of heart non linear methods are effective than linear ones A. Linear methods The time domain analysis, frequency domain analysis and geometrical analysis are categorized as linear methods. B. Time domain measures of HRV The time domain measures of HRV Consist of SDNN SDANN, RMSSD, SDSD and NN 50 Count this standard deviation route mean square values are measured in the NN interval. C. Geometrical measures of HRV:  Triangular Index: The ratio between the total NN intervals and the histogram height of all NN intervals in the time scale of milliseconds.;  Poincare plot: The R-R interval is normally plotted as an instance of the previous interval.  Differential Index: NN height intervals (at 100 and 1000 sample levels) are computed to which the histogram is applied and the differences are studied. Diagnosis of Type-2 Diabetes using Classification and Mining Techniques Sankar Padmanabhan, Manjunath K M, Madhurima V