Today I learnt about Geo position and got to know about the library used for it- Altair library. We can also use matplot for the Geo position as I have worked with both in plotting map of USA.

I have also learnt little more about clustering, specifically the “Elbow Method.” To determine how many clusters we require, I created the WCSS graph, which shows how near the points in a group are to each other. It assisted me in determining the optimal number of groups when adding another group does not significantly improve performance; we call this point the “elbow.”

WCSS (Within cluster sum of squares) is essentially a measure of how neat our clusters are. Adding the squares of all the distances between each point and the cluster’s centre. Lower WCSS results in more organised clusters. WCSS tends to decrease as the number of clusters increases. It can be used to find the right number of clusters.

In my previous post I mentioned about that there are 12 columns and today I tried to merge the csv’s on Race columns and tried to plot the graph. The graph clearly indicated that the Black people got shot is way higher than the white people got shot. But I can’t conclude with this graph that chances of black people getting shot is higher than white as there could be other factors too which resulted in black people getting shot.

I will be researching more about this and will see whether police shooting is somewhat based on race or is it some other grey area which we don’t know about.

 

I’d want to highlight the datasets based on police shootings published by The Washington Post. Every year, it is clearly stated that police in the United States shoot and murder more than 1000 people.

As I have gone through the datasets I found out that the data is from 2015 till date and it is getting updated weekly. The dataset has 12 columns: date, name, age, gender, armed, race, city, state, escape, body camera, sign of mental illness, and police department engaged.

While examining the datasets to find the key factors leading to more number of police shootings. I feel that body camera and the ethnicity is the major factor which result in more killings than other factors. This is the raw analysis just by looking at the datasets, I will deep dive under all factors in order to find a better analysis of the dataset.