Skip to content

Bike sharing is an alternative and sustainable mobility service that is always taking that is becoming increasingly popular in almost all major cities.

Notifications You must be signed in to change notification settings

NicolaRizzitello/Bike-Sharing-Analysis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

This one of my project that I did during Master's course for Statistics module. In this project I used the Random Forest algorithm to see which are the most important variables among the explanatory variable.

Language

alt text

About Dataset

The dataset was composed by 10866 rows and 14 variables initially. The data cleaning process has brought some changes so the final dataset is composed by 10944 and 11 variables:

  • date_hour: date and time of rent
  • season: season of the period (categorical)
  • holiday: working-days and no-working-days (categorical)
  • workingday: week-day and week-end (categorical)
  • weather: weather situation (categorical)
  • temp: temperature in celsius degree
  • atemp: perceived temperature
  • humidity: humidity rate
  • windspeed: wind speed
  • casual: customers not registered (response variable)
  • registred: customers registered (response variable)

About the analysis

In order to highlight differences and similarities between the habits of registered and non-registered users, an analysis exploratory allows to have a brief description of the behavior of the response variables in relation to explanatory variables.

The second step of the analysis has been to divide the dataset in training and test set, the first composed by 75% of the rows of the dataset and the test set of the 25% and I used the Regression Tree to see which are the most important explonatory variables.

About

Bike sharing is an alternative and sustainable mobility service that is always taking that is becoming increasingly popular in almost all major cities.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages