Early Detection of the Liver Disorder from Imbalance Liver Function Test Datasets
Pushpendra Kumar1, Ramjeevan Singh Thakur2
1Pushpendra Kumar, Department of Computer Applications, Maulana Azad National Institute of Technology, Bhopal (M.P), India.
2Ramjeevan Singh Thakur, Department of Computer Applications, Maulana Azad National Institute of Technology, Bhopal (M.P), India.
Manuscript received on 05 February 2019 | Revised Manuscript received on 13 February 2019 | Manuscript published on 28 February 2019 | PP: 179-186 | Volume-8 Issue-4, February 2019 | Retrieval Number: D2681028419/19©BEIESP
Open Access | Editorial and Publishing 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: Aim of this research is to develop a model for early detection of liver disorder from imbalance Liver Function Test (LFT) results’ datasets that assists the practitioners in diagnosing the liver disease efficiently. Because in the initial stage symptoms of the diseases are vague so the medical practitioners often fail to detect the disease. This study used two datasets of Liver Function Test (LFT) for building the systems, one is ILPD dataset (secondary) taken from UCI repository and second dataset (Primary) is collected form Madhya Pradesh region of India. We have used Support Vector Machine and K-Nearest Neighbour (KNN) algorithms to implement the system and Synthetic Minority Oversampling Technique (SMOTE) to balance the datasets. We have compared the results of both the algorithm on the different parameter for both the imbalanced and balanced datasets. We get the improved result for accuracy, specificity, precision, false positive rate (FPR) parameters on balanced datasets using SVM whereas using KNN we get improve results for accuracy, specificity, sensitivity, FPR and FNR parameters on balanced datasets. We can conclude that the proposed system gives the improve result on balance dataset on most of the parameter. Proposed system helps the healthcare practitioners in diagnosing the liver disease efficiently at the early stage.
Keyword: K Nearest Neighbor (KNN), Liver Function Test (LFT), SMOTE, Support Vector Machine (SVM).
Scope of the Article: Machine Learning