Big Data Processing System for Diabetes Prediction using Machine Learning Technique
B. Suvarnamukhi1, M. Seshashayee2
1B. Suvarnamukhi, Department of Computer Science, GITAM, Visakhapatnam, A.P, India
2M. Seshashayee, Department of Computer Science, GITAM, Visakhapatnam, A.P, IndiaManuscript received on September 16, 2019. | Revised Manuscript received on 24 September, 2019. | Manuscript published on October 10, 2019. | PP: 4478-4483 | Volume-8 Issue-12, October 2019. | Retrieval Number: L35151081219/2019©BEIESP | DOI: 10.35940/ijitee.L3515.1081219
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Diabetes is one of the threatening diseases to the entire mankind, though it is not fatal. Irrespective of the presence of several existing approaches for diabetes prediction, big data based diabetes prediction is quite rare. The applicability of the proposed work is wider because, medical records from different sources are extracted and the necessary attributes meant for predicting diabetes alone are processed. The goal of this work is attained by different phases such as data collection, preprocessing, attribute selection and prediction. The diabetes prediction is carried out by Extreme Learning Machine (ELM) classifier. The performance of the proposed approach is analysed by varying the classifiers and the existing approaches in terms of disease prediction accuracy, precision, recall and time consumption. From the experimental results, the efficiency of the work is proven.
Keywords: Big Data, Electronic Health Records, Machine Learning, Diabetes
Scope of the Article: Machine Learning