Active and passive portfolio management complement each other to beat the market. Fisher Black and Robert Litterman propose the Black-Litterman model (BL) as a framework for active portfolio management incorporating experts’ views in the mean-variance model. The views in BL are subjective; therefore, the method brings the opportunity to incorporate innovative and accurate technics to generate views in analytical methods or more sophisticated techniques. The recent appearance of artificial neural networks (ANN) and their variants have affected all fields. Particularly, forecasting models that use LSTM allow us to model complex relationships considering strong assumptions. This work proposes an iterative active portfolio management process that uses deep learning networks to construct the views for a BL model using tweets and fundamental variables to stocks listed in S&P 500 index. Our results notably beat the S&P500 market by 31.91% alpha annualized, namely, discounting systematic and risk-free rates.