What is the interesting outcomes from FBI gun data?
The data we used to analyze “come from the FBI’s National Instant Criminal Background Check System”.
This project is the fourth project of data sciences Udacity’s Nanodegree program.
In this project, we will follow the CRISP-DM process (Cross Industry Process for Data Mining). CRISP-DM stands for:
- Business understanding.
- Data understanding
- Data preparation
- Modelling
- Evaluation
- Deployment.
In this project, we will focus to find out:
- The highest rate of Hispanic or Latino.
- The rate of handguns in each state.
- The month which occurred the largest rate of long gun purchases in.
- The highest gun purchase for each prospective buyer.
1- What is the highest rate of Hispanic or Latino for purchasing a gun on each state?
First of all, we would like to look at the data at the state level. There is a total of 37% in California, and 30% approximately in Arizona for Hispanic or Latino buyers of guns. On the other hand, Kentucky and Alabama recorded the lowest rate of purchasing guns. Furthermore, this is not weird because of that California and Arizona have largely a population of Hispanic and Latino.
2- What is the rate of handguns against the number of states?
From the above plot, we can note that most rates of handguns occurred in 18 states. Furthermore, we can say that Florida recorded as the most state have the highest number of handguns then Pennsylvania with 51738 and 47315 respectively.
3- The month which occurred the largest rate of long gun purchases?
The given chart illustrates the month which occurred the largest rate of long gun purchases, it is suddenly increased and decreased again at the beginning. Overall, it is noticeable December showed 2335358 approximately, we can say that December contained the highest rate of long gun purchase.
4- What is the highest gun purchase for each prospective buyer?
The given bar chart illustrates the highest gun purchase for each prospective buyer, we can notice that (Hispanic or Latino) recorded the highest rate of purchasing on each state, then (Asian alone).
As can be seen from the chart, (white alone) population recorded the highest rate of the population but the least rate of gun purchasing on each state.
Important features about purchasing:
To make sense of significant features, it is important to fit a regression. Furthermore, the coefficients are telling of the impact of features.
In response, from the heatmap matrix, we can find how the type of guns, permits, and type of populations can affect the results of the total purchasing.
Conclusion:
In this blog post, we discovered FBI gun dataset from Github and got some interesting outcomes.
We found that populations type play a crucial part in a purchase’s total performance, also types of guns are making sense of significant changing purchasing each year and month. Furthermore, There were quite a few states did not affect by a total of purchasing.
For finding more about this analysis, take a look at this Github link.