Machine learning techniques facilitate efficient analysis of complex networks, and can be used to discover communities. This study aimed use such approaches to raise awareness of the COVID-19. In this regard, social network analysis describes the clustering and classification processes for detecting communities. The background of this paper analyzed the geographical distribution of Tambaram, Chennai, and its public health care units. This study assessed the spatial distribution and presence of spatiotemporal clustering of public health care units in different geographical settings over four months in the Tambaram zone. To partition a homophily synthetic network of 100 nodes into clusters, an empirical evaluation of two search strategies was conducted for all IDs centrality of linkage is same. First, we analyzed the spatial information between the nodes for segmenting the sparse graph of the groups. Bipartite The structure of the sociograms 1-50 and 51-100 was taken into account while segmentation and divide them is based on the clustering coefficient values. The result of the cohesive block yielded 5.86 density values for cluster two, which received a percentage of 74.2. This research objective indicates that sub-communities have better access to influence, which might be leveraged to appropriately share information with the public could be used in the sharing of information accurately with the public.
Keywords: COVID-19 community; clustering; node metrics; social network.