Electricity Billing and Consumption Prediction using Machine Learning

This research suggests a AI-Learning based method for forecasting electricity usage and bill amounts. The project develops and assesses a number of machine learning models using historical electricity use data and customer billing information. The dataset, which consists of several months' worth of historical data from a residential power client, is used to train and test the models. The study shows that the suggested machine learning models are highly accurate in forecasting electricity consumption and bill amounts. According to the results, machine learning-based methods can greatly increase the precision of predictions of power use and bills, which can be helpful for both consumers and utility providers. The goal of this paper is to create a system that fuses the findings of various machine learning techniques, in order to conduct accurate prediction for a tenant. K nearest neighbour, Support vector machine and Artificial Neural Networks are few of the algorithms that are employed. Each algorithm's accuracy is determined along with the model's accuracy. The model for predicting is then chosen from those with decent accuracy. The dataset used for building the model is Smart metres Database, which was selected from Kaggle website.