The mapping of crime incidents in the spatio-temporal domain is one of the vital components to take up more decisive steps to minimize crime incidents. Distribution of crime incidents is not accidental as they occur due to the various socio-economic as well as environmental conditions. It has been reported that crime rates are rising frighteningly in developing countries. In India, several studies on the analysis of crime patterns, trends, and the causes have been carried out with the inputs from geospatial technology. There is no detailed study on the analysis of crime hotspots using spatio-statistical techniques for the Shillong city of Meghalaya, India. The main objective of this work is to determine the pattern of crime hotspots during different time frames for the Shillong city using the incremental spatial autocorrelation method. Three years of crime incident data (2014-2016) comprising of more than 4515 incidents were analyzed and categorized mainly in 5 major types such as house trespassing, murder, crimes against women, theft, and cheating related charges. The intensity of the spatial clustering of the crime incidents were obtained effectively by the z-score measured by increasing the distance band or threshold distance, which helped to understand the spatial distribution of crime incidents with a higher degree of confidence level. It was observed that the hotspots were primarily confined to the central part of the city and some of the hotspots (average) were unevenly distributed over space irrespective of the size of the population density. The crime against women were reported mostly during the evening and the highest theft were reported after 2200 hrs during night. Other crime cases were not observed with any specific pattern concerning the time of occurrences.