Liver disease diagnosis using predictive analytics-based machine learning models

In the world, liver diseases cause roughly one million deaths. Several traditional methods can be used to diagnose liver issues, but they are expensive. Early prediction and treatment of liver disease may benefit everyone at risk. Due to its early illness prediction capabilities, machine learning has a significant impact on healthcare as technology develops. This study assesses how well machine learning predicts liver disease. This article introduces the liver disease prediction (LDP) approach, which enables researchers, stakeholders, students, and medical professionals to predict liver illness. However, classification machine learning models such as logistic regression (LR), support vector machines (SVM), decision trees (DT), and random forests (RF) were used in this study. Python uses accuracy comparison to forecast the outcome. The random forest algorithm predicts liver diseases with the highest accuracy, per the results. Above the permitted accuracy threshold and could be taken into account when determining the prognosis of liver disease. It adjusts the learning process's cost function to account for the unequal distribution of classes, improving the model's performance. To predict liver disease, we first acquire an evenly distributed dataset and then train a machine learning model (specifically, logistic regression, support vector classifier, and gradient boosting classifier). A different test dataset is used to evaluate the effectiveness of the suggested model using a variety of criteria, including accuracy, precision, recall, and F1-score. This model is expected to significantly aid medical professionals in the field by effectively addressing class imbalance through data balancing algorithms.