Multivariate Time Series Forecasting with LSTM for Madrid, Spain pollution

Time series forecasting is something of a dark horse in the field of data science and it is most critical factor that decides whether a business, temperatures or any environmental factors effect will rise or fall, A single time-dependent variable means A univariate time series while A Multivariate time series like environmental data has more than one time-dependent variable. Each variable depends on its past values and also on other variables past values. this paper used Neural networks like Long Short-Term Memory (LSTM) for forecasting Spain capital Madrid Air Quality using a dataset that reports on the weather and the level of pollution each hour for two years from 2015to 2016 the data includes the date-time, the pollution concentration of the following: SO2, NO_2, NO, CO. Almost the best problems modelling for multiple input variables are recurrent neural networks and they are the great solution for multiple input time series forecasting problems, where classical linear methods can't. this paper used LSTM model for multivariate time series forecasting in the Keras and Tensor Flow deep learning library in a Python SciPy environment with Machine Learning scikit-learn, Pandas, NumPy and Matplotlib libraries.