New York's energy bill fluctuations are a growing concern of the people. Due to the unpredictable changes in energy demand, suppliers struggle to allocate sufficient resources for energy production. This imbalance causes energy bill rates to surge which leads to residents not being able to plan their household budgets accordingly. The goal of this project is to build a robust Time Series Forecasting Model that predicts Energy Demand effectively. In addition to this, the project also attempts to investigate the reason behind the anomalous highs and lows in energy demand by studying social media sentiments. Data for Energy Demand is collected from New York ISO website. ARIMA model is built for Time Series Forecasting. Twitter Data is streamed for anomalous data and topic Modeling techniques like LDA and GSDMM are used for the analysis of textual data.
