In today’s era medical information and disease detection as well as health care depends greatly on machine learning. Modern machine learning algorithm’s make an advantage in detecting the big health hazards like Brain Tumors, Cancer, Covid 19, Diabetes, Kidney Diseases etc. Kidneys are the most vital organs in human body, if any issue occurs in kidney it can have harmful effect on body. Kidney sickness comes in two different forms: Acute Kidney Disease and Chronic (long-term) Kidney Disease and this must be emphasized. Acute kidney disease is the most prevalent kind of kidney disease. Kidney function gradually declines over years, which is the hallmark of kidney failure. Chronic Kidney Disease is a dangerous condition that can be extremely fatal if not identified and treated at its early stages. Thus, the primary objective of the present work is to use machine learning techniques to accurately forecast Chronic Kidney Disease. For this, clinical dataset of 400 patients with 25 attributes related to this disease is obtained from the well-known UCI ML repository. Since the dataset is large and potentially expensive modeling costs, information gain is also utilized in the feature selection process. However, dataset has numerous of missing variables. To obtain a clean dataset, pre-processing is carried out. After pre-processing, three different machine learning approaches Decision Tree, Logistic Regression and Random Forest are applied on dataset. Then result of each three approaches are compared using a range of performance indicators. According to the experimental findings Decision Tree, Logistic Regression and Random Forest achieved an accuracy of 97.5%,97.5% and 98.75% respectively in disease prediction. Regarding the ability to forecast Chronic Kidney Disease, Random Forest is thought to be superior than others.