To start, hotspots are areas of higher concentrations of data points, such as particular neighborhoods where the crime rate is abnormally high or swaths of the country that are impacted by an above-average number of tornadoes. Hotspot analysis is the process of finding these hotspots, should any exist, in a population using sampled data. This process is generally done by leveraging kernel density estimation.
Hotspot analysis can be described in four high-level steps:
Collect the data: The data should include the locations of the objects or events. As we have briefly mentioned, the amount of data needed to run and achieve actionable results is relatively flexible. The optimal state is to have a sample dataset that is representative of the population.
Identify the base map: The next step is to identify which base map would best suit the analytical and presentational needs of the project. On this base map, the results of the model will be overlaid, so that the locations of the...