Seoul Bike Rental Exploratory Data Analysis

Image by Freepik

Motivation

• Kaggle Educational Competition : link
  1. You are provided hourly rental data along with weather data.
  2. For this competition, the training set is comprised of the first 20 days of each month, while the test set is the 21th to the end of the month.
  3. You must predict the total count of bikes rented during each hour covered by the test set, using only information available prior to the rental period.
• Problem :
  1. Forecast total count of bikes rented.
  2. Use classical machine learning models (no deep learning models).
• Solution :
  1. Data analysis.
  2. Feature Engineering.
  3. Modeling.
  4. Forecast.

Note: there is an account of the time-series nature of the data that was not described in the EDA provided and was only implemented in the feature engineering phase.