ABSTRACT With the increase in the number of patients with liver disorder it has become a fatal disease in many countries. These disorders of liver need clinical care by professionals in healthcare. Data mining provides with essential methodology for computing applications in the domain of medicine. Medical data mining uses a set of approaches that extract valuable patterns from the human services databases to help doctors pick the best diagnosis. Many classifiers had been used to predict the liver disorder. Hybrid algorithms for data mining are a logical combination of multiple pre-existing techniques to enhance performance and provide better results. In the proposed work, a hybrid algorithm is introduced which uses the concept of clustering and classification for prediction analysis. We have analyzed with real patient dataset for constructing hybrid approach to predict liver disorder taken from the UCI repository. The algorithms used in this work are K-means clustering, Random Forest classifier. This hybrid algorithm is implemented using Jupyter Notebook. This approach is compared with existing classifiers like KNN, LR etc using the confusion matrix parameters such as precision, recall, f1-score and accuracy. The hybrid approach shows better results than other algorithms for prediction analysis of liver disorder.