Proceedings of the First International Conference on Advanced Scientific Innovation in Science, Engineering and Technology, ICASISET 2020, 16-17 May 2020, Chennai, India

Research Article

Stock Trend Prediction using Deep Neural Networks in Time Series and Social Sentiment Analysis

Download240 downloads
  • @INPROCEEDINGS{10.4108/eai.16-5-2020.2304210,
        author={P.  Ravichandran and J. Dafni Rose and K.  Vijayakumar},
        title={Stock Trend Prediction using Deep Neural Networks in Time Series and Social Sentiment Analysis},
        proceedings={Proceedings of the First  International Conference on Advanced Scientific Innovation in Science, Engineering and Technology, ICASISET 2020, 16-17 May 2020, Chennai, India},
        publisher={EAI},
        proceedings_a={ICASISET},
        year={2021},
        month={1},
        keywords={neural networks word2vec lstm},
        doi={10.4108/eai.16-5-2020.2304210}
    }
    
  • P. Ravichandran
    J. Dafni Rose
    K. Vijayakumar
    Year: 2021
    Stock Trend Prediction using Deep Neural Networks in Time Series and Social Sentiment Analysis
    ICASISET
    EAI
    DOI: 10.4108/eai.16-5-2020.2304210
P. Ravichandran1,*, J. Dafni Rose1, K. Vijayakumar1
  • 1: St.Joseph’s Institute of Technology, OMR, Chennai-119
*Contact email: ravichan3064@gmail.com

Abstract

In this paper, we propose an approach towards predicting the trend of stock price values by analyzing the relevant words occurring in social media like Twitter and by performing a time series analysis of the performance of the stock over the years. We obtain training data and train them separately against normalized values of stock prices themselves using neural networks and obtain the desired results by using the outputs of these separate approaches as the training data for another separate neural network that predicts the trend in the stock’s future pricing along with the values with a certain degree of accuracy.