Step I :

As said before most of these low profile activities are not even recorded but there are some which go through lawsuits, and WPRDC has this data recorded and up on their site. Thus I use this data to explore the areas which face the most number of crimes and what is the type of crime. What age, sex and race these criminals belong to so, as finding ways to educate them and also make people aware of these people. 

The WPRDC dataset, consist of the criminal activities recorded by Pittsburgh Police from 2016-2017.

Datasets Used : 

1) Non-Traffic Citation, Pittsburgh - https://data.wprdc.org/dataset/non-traffic-citations

2) Police Blotter Data - https://data.wprdc.org/dataset/uniform-crime-reporting-data

3) Pittsburgh Neighbourhood Shapefile - https://data.wprdc.org/dataset/pittsburgh-neighborhoods1

As the number of crimes listed in the data set was a broad range, grouping the similar kinds of crime into in the dataset was a challenge. While sorting the dataset, I realized that these crimes locations were most nearby to each other, thus in order to check, I used the Clustering analysis to visualize the areas most affected.