CLUSTERING: ANALYSIS OF ALGORITHMS AND APPLICATIONS

Abstract

With the rise of the application of Machine learning in academia and industrial sector, clustering has become an important field of study. Clustering has been extensively used in studies involving unlabeled data, image processing and unsupervised learning. We have taken up the concept of clustering for our study and we have discussed the performance of two very popular and well used clustering algorithms and the application of clustering. The application of clustering has been discussed in the context of COVID 19 and medical field. In the First part, we compare K-Means and BIRCH Clustering algorithms on multiple datasets, and derive our results. After considering those results, we would move to discuss the application part. As we know, the world is witnessing an unprecedented catastrophe as a result of the COVID-19 epidemic, which has spread to approximately 216 nations and territories throughout the globe. A COVID-19 infection may progress to pneumonia, which may be diagnosed by CXR (Chest X-Ray) examination and should be treated as soon as possible after diagnosis. This part would be intended to examine the use of artificial intelligence in speedy & accurate diagnosis of COVID-19 pneumonia utilizing digital CXR pictures. In this research, we use a machine learning (ML) method i.e. SVM (Support Vector Machine) classification technique. SVM was used in the development of the model. The purpose of this research has been to use clustering, image processing, image segmentation, and feature extraction in fast or accurate identification of COVID19 chest X-ray or CT images. We assessed the performance of ML techniques on chest X-ray pictures as well as CT scans to COVID-19 diagnosis in this work. The model's performance was assessed using relevant classification measures, such as accuracy, precision, recall, & F1 score, among others.