Primary Energy (PE) Consumption 🌍

The objective of this project is to assess the suitability of an engineering dataset for machine learning modeling, determine data features that should be used for machine learning, and then fit and critique the performance of ML model(s) using uncertainty quantification. This project is part of the Engineering Applications of Machine Learning course at Georgia Tech.

Objectives 🎯

After some Exploratory Data Analysis (see EDA), I have listed 2 possible ojectives for this project:

  • Predicting the CO2 emissions
  • Predicting the primary energy consumption per capita

Many projects are predicting the CO2 emissions, so I will focus on the second objective. That’s why in the modeling part of the project, I will focus on predicting the primary energy consumption per capita.

Projects Steps 💡

This project is divided in 3 main parts that are done in the following order:

  1. Exploratory Data Analysis (EDA): critical step in the data analysis process where the main goal is to summarize the main characteristics of a dataset, often with the help of graphical representations

  2. Feature Engineering: the aim of this part is to improve the future machine learning training (better performance and greater accuracy) by selecting relevant features, handling missing data, etc…

  3. Modeling: The modeling part of a machine learning project is a crucial phase where the actual machine learning algorithms are selected, trained, and evaluated.