Chronic Kidney Disease (CKD) poses a substantial global health challenge, characterized by elevated morbidity and mortality rates and the potential to trigger cascading health issues. The mild beginning of CKD, characterized by the absence of overt symptoms in the early stages, frequently results in a lack of awareness among patients. Timely detection of CKD is crucial for providing patients with prompt interventions to mitigate disease progression. In this paper, we have analyzed performance of six different algorithms (J48, Random forest, Naive Bayes, Support Vector Machine, MLP, and K-Nearest Neighbor) on the CKD dataset from Kaggle.com and provided a performance analysis using the evaluation parameters. A thorough review of the literature on machine learning models for CKD early detection is also provided in this research. We categorize and examine several machine learning approaches, including algorithms, datasets, and assessment metrics used in CKD prediction, after conducting a thorough analysis of the current literature. Our study reveals essential techniques and outcomes, providing insight into the complexities of current strategy. We address the clinical implications of these findings and identify potential areas for future investigation. The results obtained on the six algorithms shows that SVM has achieved the highest accuracy of 97%.