Predicting liver disorder based on machine learning models

As the main detoxification organ of human body, liver is very important in humans' health by metabolizing a lot substances that are taken in, including alcohol and medicine. However, if person consumes too much or contaminated food, it will lead to disorder causing little ingestion essential nutrients. Accurate prediction for consumption, therefore, providing doctors necessary information diagnosing diseases. To address this problem, paper introduces machine learning models predict disorder. In addition, alleviate influence data randomness splitting set into training testing set, leave-one-out cross valuation utilized. The feature importance relationships between different features also analyzed. experimental results showed effective consumption prediction. Among them, random forest has best performance terms accuracy (80.35%). reason could be ensemble strategy used helpful reduce over-fitting problem caused imbalanced set. This indicates useful tool suggestions