Data Exploration

Case Study: Altoona Crime Rates

Imagine this: You have been asked by the City of Altoona’s police department to help them improve their ongoing fight against crime in the city. They provide you with Altoona Crime Rates, a dataset covering their arrests over the last few years, and ask you to provide as many insights and recommendation as you can. They also provide you with Altoona Population Estimates, a dataset covering the city’s population, to help in your analysis. The following 10 Modules show the steps you can do as their Data Scientist to help them make their law enforcement more efficient.

#	Module	Explanation
1	Accessing Data	Import data on Altoona crime rates into the repository. Add data to the process. Inspect data using summary statistics.
2	Filtering & Sorting	Determine the most common crimes committed by juvenile offenders.
3	Merging & Grouping	Join the Altoona crime dataset with the Altoona population dataset to compare crime statistics to general population statistics in the area.
4	Creating & Removing Columns	Calculate percentage of Altoona population who are convicted offenders, in general, by race, and by ethnicity. Keep only interesting columns.
5	Changing Types & Roles for Modeling	Prepare data for modeling by changing column types. Predict the profile of Altoona criminals most likely to commit a crime based on sex, age, race, and ethnicity.
6	Normalization and Detecting Outliers	Normalize the Altoona crime data, and remove outliers to improve the predictive model’s performance.
7	Pivoting and Advanced Renaming	Pivot the Altoona crime data from long table to wide table format, then use advanced renaming of the attributes to make the final outcome neater.
8	Handle Missing Values	Perform data cleansing to achieve higher data quality.
9	Macros & Sampling	Use macros to calculate a new example set size; then sample the data down to this new size.
10	Looping & Branching	Loop over the 2 sexes and sample the examples for each sex individually down to 500 or less.

Datasets (download here)