Random Search Optimization Using Random Forest Algorithm For Liver Disease Prediction

Abstract

The liver is a vital human organ with complex and diverse functions. One of the diseases that affect the liver is hepatitis or liver disease. Early detection is crucial to enable more effective intervention and slow the progression of the disease. However, diagnosing liver disease often faces challenges, especially in detecting the early stages of the disease from complex and diverse medical data. This study aims to optimize the Random Forest algorithm using the Random Search method for liver disease detection. The Random Forest algorithm is applied as the primary model in this research, while hyperparameter optimization is performed using the Random Search method to enhance model performance. The results show that the Random Forest model without optimization achieves an accuracy of 93%. After hyperparameter optimization, the model's accuracy increases to 94%. In conclusion, applying hyperparameter optimization using the Random Search method successfully improves the performance of the Random Forest model. The resulting model provides more accurate predictions.