Data Transformations and GIS Analysis Practical Example

In this tutorial, you will learn how to make use of Pandas library of python programming language to transform a dataset using various data wrangling and manipulation techniques, and then use that dataset in QGIS for performing geospatial analysis.
In this example, we download a dataset from the UK police database, which shows crime incidents recorded in the city of London. Usually the freshly downloaded data in its raw format are not in a suitable structure to be used for analyses, hence will require undergoing a number of intermediary data transformation steps.
We make use of the Python Pandas library, and upon reading the csv files, we record them into separate Pandas DataFrames, and after that we use pandas.concat to merge them all into one large dataset. We also perform some data cleaning work by getting rid of some columns that are deemed unimportant for our analysis, using the same library.
Finally, we import the cleaned dataset into QGIS using latitude and longitude information provided, to visualize the spatial locations of the crime incidents. Furthermore, by way of employing a categorical symbology type, we also visualize the crime incidents by type, which can provide a much clearer outlook of the data. We also create a heatmap based on the density of the points, which allows us to clearly visualize where the crime hotspots are located at. We use QGIS for performing all of the spatial visualization tasks, including the creation of the spatial heatmaps for identifying crime hotspots.
#dataanalysis #spatialdata #geodeltalabs