Data Science Project | Bike Rental Count | Part 1 | Missing Data Analysis

DATA SCIENCE WISDOM COMES ONLY THROUGH FAILED EXPERIMENTATION by Damian Mingle

Welcome to Bike Rental Count Project!

The objective of this project is to learn how to build a project from scratch till the end.

The aim is to predict the count of bike rentals based on the seasonal and environmental settings. By achieving this goal, it would be possible to help accommodate in managing the number of bikes required on a daily basis, and providing better services to its customer.

In the Project series, I will be providing the codes and explain them step by step. Our focus is on Quality of codes not on quantity of codes.

Bike Rental Startup Bounce

First, Set your path to the desired location using setwd command where the file is located. Do write the name of the file properly and check the extension of the file in the properties of the file to be sure.
The file that i am using is in CSV format and each column is separated by comma. So, that's why i used the argument "sep" in the read.csv line and also mentioned the separator that is comma.

Now, our file fetched into the object name "bike". We will use "bike" from now onward for our further implementations.

"str" is a compact way to display the structure of an R object. It will output the information on one line for each basic structure.
In simple meaning, if we want to know the number of observations and variables in the file. It's one of the command you can look up to. Also, It will display the type of each variable like First variable is integer, Second is factor etc.
It is a very important command.

output of str command

Next, we can use "describe" command to understand more about our data. If you are unable to understand it, it would become a difficult task for you in further implementations. You will not have any clue which command have to use.

"describe" is provided by "psych" package. You have to install the package to use it.

Output of  "describe" command-

Output of describe command

Basically, this command would give you a descriptive summary. We can see that our data is explained by 13 statistics term. I will explain this command with all its term in deep in other separate post.

"colnames" command will only give you the list of all the columns(variables) in your data.

Output of colnames command

Missing Data Analysis

"missmap" command is one of the most used command and personal favorite of mine. There are multiple ways for finding out missing data.
 It is found under the package "Amelia".

Output of missmap command

If there is any missing data in particular variable, it will appear in red color in the graph.

Converting required Variables into factor

Replacing Categorical variable value by meaningful information

There is a lot more in this project. Next, we will do Exploratory Data Analysis(EDA) in second post of this project.


Feel free to comment your reactions!💓.