In Fog computing environment the major issue is task scheduling and resource allocation. The study tries to address the above-mentioned issue using clustering machine learning algorithms like Canopy Clustering, Hierarchical Clustering and Make Density Based Clustering. The major objectives being associated with the research work were firstly to do comparative analysis of various clustering algorithms based on performance measures overall accuracy and to identify the most suitable clustering algorithm for resource allocation and task scheduling. The hypothesis testing results confirms that there is significant difference between various clustering algorithms based on performance measure overall accuracy at the overall accuracy values varies highly, the accuracy value of Canopy Clustering, Hierarchical Clustering and Make Density Based Clustering were found to be 74.5%, 59% and 48.5% respectively. Among all the clustering algorithms the most appropriate one for task scheduling and resource allocation is found to be Canopy Clustering.