Machine Learning Based Analysis for Liver Disorder Prediction

Abstract

The liver plays a prime part in the body generating proteins and blood condensing to soaked fat, sugar, and all other digestion. It has a variety of purposes, counting and eliminating poisons from the body, and is vital to happen. It holds an extensive diversity of purposes which comprises Preliminary estimate of any sickness is very noteworthy to defend human natural life and yield suitable stages to evade the sickness. AI approaches stay suggestively actively busy in foremost categories of therapeutic knowledge. This project work comprises initial enquiry and forecast of liver complaint by means of plentiful machine learning and deep learning methods. The developed model is examined by various performance measures such as accuracy, recall and precision. A comparative analysis was carried out by using some methods such as logistic regression, naive Bayes. We optimised the method using ensemble methods such as random forest and XG boost classifiers. Finally, we applied neural networks for getting maximum accuracy of 95%.

Keywords:  Liver Disease, Machine Learning, Deep Learning, Accuracy.