kanishkparganiha / classification-of-different-building-energy-score-using-machine-learning-and-deep-learning-method

Classification-of-different-Building-Energy-score-using-Machine-Learning-and-Deep-Learning-method

I. Background and Introduction

⚫ Introduction In recent years, people's attention to the problem of energy shortage has increased significantly. To solve the energy over-utilization problem, government or energy company always renew their policy or cost method to reduce the waste of energy. But the problem is how to define energy waste and how to make a good policy to limit users who waste resources without harming ordinary users. This project can provide a possible solution to the above problems.

⚫ Problem definition Of the 25 variables provided by the data, we need to summarize a reliable parameter that represents the energy and water use of each building. Based on this parameter, we can score each user and determine whether this user has overused the resources. Based on the overall resource forecast and usage, we can help the government or company define new policies or pricing methods to constrain the situation of resource waste or over-utilization. As stated in the report that we are building a predictive model that correlates the energy data to the property use details to identify the key drivers of energy use and predicts the Energy Star Score which is a measure of how well a property is performing relative to similar properties when normalized for climate and operational characteristics. The 1-100 scale is set so that 1 represents the worst-performing buildings, and 100 represents the best performing buildings. A score of 50 indicates that a building is performing at the national median, taking into account its size, location, and operating parameters. The goal