Ensemble Mixed Breed Deep Clustering Algorithm for Complex Datasets
S. Nagarjuna Reddy · International Journal for Research in Applied Science and Engineering Technology · 2019
Clustering is process of having similar items at one place. Many datasets are available for the research in clustering. Machine learning (ML) and deep learning (DL) are two latest domains that to improve the clustering techniques. From past decade so many applications are developed in clustering for pattern recognition, speech and other prediction type of algorithms. According to the latest research, deep clustering algorithms can be used to learn better representations of the data. In this paper, the Ensemble Mixed Breed Deep Clustering Algorithm (EMBDCA) which is adopted various deep learning algorithms for improving the performance. For the training, Information Maximizing Self-Augmented Training (IMSAT) is utilized. This will improve the accuracy especially for the datasets such as mushroom and MIST dataset. The parameters sensitivity, specificity and quality of clusters are also improved.