.png){kind=link}
Your company has a project to analyze and create predictive models for the fuel economy of cars. Fuel economy data are the result of vehicle testing done at the Environmental Protection Agency's National Vehicle and Fuel Emissions Laboratory in Ann Arbor, Michigan, and by vehicle manufacturers with oversight by the EPA. Analysis of the data from 2008 to 2018 has to be done. You can download the data using the EPA Green Vehicle Guide - Datafile3 and the data description
You, as part of the project, have to explore, analyze and treat the data for the below-mentioned points.
- Number of samples in each year
- Number of columns in each dataset
- Duplicate rows in each dataset
- Data types of columns
- Features with missing values
- Number of non-null unique values for features in each dataset
- What those unique values are and count for each
- Analysis by grouping certain features
Draw certain conclusions by creating visuals to communicate the change in the cars over 10 years.
Example:
- Are there more unique models using alternative fuels in 2018 compared to other years? By how much?
- How much have vehicle classes improved in fuel economy (increased in mpg) per year?
- What are the characteristics of SmartWay vehicles? Have they changed over time? (mpg, greenhouse gas)
- What features are associated with better fuel economy (mpg)?
Also, you need to merge all the datasets with the same column names by adding a column of year to differentiate the data for different years (drop the extra columns. Also, make sure that the data types of merged columns are the same.
Find the appropriate number of clusters from the data and conclude the properties of the clusters.
Create a Predictive model to predict whether a vehicle is SmartWay vehicle or not using the Classification Algorithm (Logistic Regression and Decision Tree).
Also, conclude which is the best model and why not the other.
Also, create a model to predict the City Mileage per gallon (City MPG) using the best features.
- Data Source: [https://www.fueleconomy.gov/feg/download.shtml ( EPA Green Vehicle Guide - Datafile3 )]
- Data Description: [https://www.fueleconomy.gov/feg/EPAGreenGuide/GreenVehicleGuideDocumentation.pdf]
- Anaconda - version 4.2.0 or higher (using Python version 3.5.2 or higher) downloaded on your system ( from the software house)
- Python Packages Required: Numpy, Pandas, Matplotlib, Scikit-Learn (sklearn)
- [https://lex.infosysapps.com/toc/lex_16286352104288236000]
- [http://pythonplot.com/]
- [http://www.sthda.com/english/articles/29-cluster-validation-essentials/96-determining-the-optimal-number-of-clusters-3-must-know-methods/]
Use the guidelines below for getting the project evaluated by your SMEs from your project or account.
Component Marks:
- Data Analysis and cleaning for all 10 datasets = 15
- Data visualization – at least 5 plots per dataset = 20
- Correct merging all the datasets with an additional column, year = 15
- Cluster creation with the appropriate number of clusters and conclusion = 15
- Appropriate classification Model (2 models) with accuracy score = 20
- Linear Regression Model to predict City MPG = 15
