Stock Index Prediction with Financial News Sentiments and Technical Indicators

The price of a stock in the market can be influenced by many factors, out which sentiment investors plays vital role. Most often, depends on news headlines. Therefore, headlines also play an important role fluctuation index. This paper uses combination Bidirectional Encoder Representations from Transformers (BERT) and Gated Recurrent Unit (BiGRU) algorithms for prediction scores based national financial data. Technical indicators like Relative Strength Index (RSI), Exponential Moving Average (EMA), Convergence Divergence (MACD), Stochastic Oscillator along with normal ’Date’, ’Open’, ’Close’, ’High’, ’Low’ ’Volume’ data used to predict short-term momentum value. BiGRU algorithm index value (a) technical only (b) scores. Keeping all hyperparameters constant, provided better results when were added dataset parameters as input.