Empirical Study of Semi-Supervised Deep Fuzzy C-Mean Clustering Algorithm

Nowadays machine learning is most widely used in different applications to enhance the quality of system. In our study, we are going to propose an Empirical study of semi supervised Deep Fuzzy C-Mean clustering algorithm on different state-of-the-art pre-processing approach. However, the performance of the models depends on the quality of datasets. In this paper, we initially split our data into labeled and unlabeled data and simultaneously compare the feature between labeled and unlabeled data to extract the knowledge from unlabeled data. In second stage of feature selection and instance reduction, we apply Information Gain (IG) to conduct redundancy control and Random under sampling and Random over sampling to handle the imbalance problem. In our paper, we use the dataset (MNIST) to check the demonstration performance of our given approaches.