A Comparative study of different approaches for clustering of incomplete medical data

Fuzzy c means is a conventional clustering algorithm that uses complete data sets for the clustering process, making it difficult to deal with incomplete data which is a critical problem in medical research that cannot be avoided. These missing attributes are due to several factors. A variety of imputation and non-imputation-based methods are used to estimate missing data. We reviewed various clustering algorithms used to deal with missing data in the medical sciences in this research work. The four most commonly used non-imputation and iterative clustering strategies and a variety of imputation-based FCM clustering algorithms are thoroughly examined. The Thyroid and Wisconsin breast cancer dataset naturally contain missing values from the UCI repository that are chosen for experimental results.