Cirrhosis Prediction Using Efficient K-nearest Neighbour Machine Learning Algorithm
Main Article Content
Abstract
Machine learning has become a crucial tool in data engineering, particularly for predicting outcomes from existing datasets. Classification, one of the most widely used machine learning techniques, involves learning patterns from a dataset and applying them to predict outcomes in new data. However, many traditional classification algorithms face challenges with limited accuracy. This study introduces a novel approach, the Supervised Learning Technique, which surpasses the performance of existing algorithms such as Logistic Regression, SVM, Decision Trees, Naive Bayes, and Random Forests. Using a K-Nearest Neighbors (KNN) classifier, the proposed method achieved an impressive 93% accuracy in disease classification. To make the results accessible, the researchers developed a user-friendly web application that allows users to input data and receive predictions. The Indian liver disease dataset was used to demonstrate how this technique can significantly enhance prediction accuracy for liver disease. The primary goal of this study is not only to improve classification accuracy but also to highlight the algorithm's potential for early-stage disease detection.