Multi-Linear Regression Technique for Predicting the Liver Disease

The liver is one of the fundamental organs in the human body, and it gets affected for several reasons. The various strategies to distinguish liver illness incorporate invasive and non-invasive ways. The outcomes attained using invasive techniques are indecisive. The liver biopsy is prone to sampling error, variability in the results given by different clinicians, pain occurring during the biopsy, higher cost of carrying out this process, and drawbacks like the patients with some other medical complications may respond negatively. Therefore, an accurate prognosis is significant. Another preponderant way to diagnose a disease includes the use of Machine Learning (ML) algorithms to analyze the abnormality. Soft computing and machine learning have become popular in generating promising solutions to the processing, predicting, and visualizing of the data received from the Internet of Things. Machine Learning Techniques are very much suited for risk modeling and management and best suited for making any decisions and generating a decision support system. Using the domain ‘machine-learning,’ a decision-based health care system can be created quickly. So, at this moment, we implement this paper by using machine learning techniques like decision trees, multilinear regression, genetic algorithms, and random forest algorithm to predict liver disease. This paper focuses on using the above-mentioned pre-existing algorithm and gives prominence to the possibility of improving the same.