/Stock-Price-Prediction-using-numerical-and-text-data

Stock Price Prediction using the historical stock closing price data and news headlines data of the the stock.

Primary LanguageJupyter Notebook

Stock Price Prediction using numerical and text data

Python 3.7

Objective

The objective of this project is to forecast the closing price of a stock based on the historical data of the stock and the news headlines of the chosen stock.

Directory Tree

├── lexicon_data
├── models
├── historical_data.csv 
├── Hybrid Model.png
├── MinMaxScaler
├── model.h5
├── model.json
├── README.md
├── Sentiment Analyzer
├── Stock Market Prediction using Numerical and Textual Analysis.ipynb

Dataset

The numerical dataset of BSE Sensex is taken from Yahoo Finance. This dataset contains last 15 years of stock details of BSE Sensex. The dataset link is given here. The text dataset is taken from Harvard dataverse. This dataset contains the last 20 years news headlines of different categories published in Times of India. The dataset link is given here.

Model Structure

The whole hybrid model which combines the insights from text and numerical data is represented below:

Code

The code for the model, algorithms used and accuracy of the model can be found here. The trained model weights is available here.