The liver, the body’s greatest inner tissue, is one of the significant diseases that, for several causes, impair an individual’s usual, healthful proportions. Fatty liver, cirrhosis, hepatitis, chronic liver disease, liver cancer, and liver tumors are just a few of the many disorders that may harm the liver. Excessive triglyceride accumulation of fat results in fatty liver. Machine learning algorithms are designed in this study by investigating with various data balancing methods, including SMOTE + Encoded Nominal and Continuous (ENC), SMOTE + TOMEK, Synthetic Minority Oversampling Technique (SMOTE), KMEANS SMOTE, SMOTE + Edited Nearest Neighborhood (ENN), and SVM SMOTE, with Random Forest providing the best precision with SEMOT-ENN balancing technique. The collection of data is also cleaned, and the number of characteristics is decreased using tilt and correlate techniques. Subsequently to standardize characteristics, the PCA (principal component analysis) algorithm is applied. Lastly, the Random Forest classifier using the SEMOTE-ENN balance algorithm performs the prediction function. The Liver Disease information set from the KAGGLE or UCI websites has been used. The effectiveness of every approach is assessed based on its exactness, precision, and complexity matrix.