Leverage Financial News to Predict Stock Price Movements Using Word Embeddings and Deep Neural Networks

By Yang Peng, Hui Jiang. 2015

90 citations recorded by the source corpus.

Financial news contains useful information on public companies and the market. In this paper we apply the popular word embedding methods and deep neural networks to leverage financial news to predict stock price movements in the market. Experimental results have shown that our proposed methods are simple but very effective, which can significantly improve the stock prediction accuracy on a standard financial database over the baseline system using only the historical price information.

Subfields: Electrical and Electronic Engineering, Finance, Management Science and Operations Research, Natural Language Processing, Information Extraction

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Electrical and Electronic Engineering

In Finance

In Management Science and Operations Research

In Natural Language Processing

In Information Extraction

Citations in the corpus

Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019 (2019, methodology): “[202] News from Reuters and Bloomberg, Historical stock security data 2006-2013 News, price data 5d 1d DNN Accuracy -”
Deep Learning for Financial Applications : A Survey (2020, methodology): “In [152], text mining and word embeddings were used to extract information from the financial news from Reuters and Bloomberg to predict the stock price movements.”
Deep Learning Approach for Short-Term Stock Trends Prediction Based on Two-Stream Gated Recurrent Unit Network (2018, methodology): “Because of its efficiency on large-scale datasets [17], researchers have already applied some DNN models on features extracted from news articles and historical stock prices such as [18] and [19].”
The Impact of Structured Event Embeddings on Scalable Stock Forecasting Models (2015, methodology): “Using the same dataset, Peng and Jiang [16] also adopted structured events to analyse market movement polarity with deep neural network and correlations between companies.”
Intraday prediction of Borsa Istanbul using convolutional neural networks and feature correlations (2017, methodology): “Proven performances in these tasks encourages researchers to use them in stock market prediction [7, 8].”
DeepClue: Visual Interpretation of Text-Based Deep Stock Prediction (2019, background): “Previous work has demonstrated an over 60 percent accuracy in predicting the daily stock pricemovement using deep neural networks over a large collection of financial news [5], [6], [7].”
A New Model for Stock Price Movements Prediction Using Deep Neural Network (2017, methodology): “We split the dataset into three sections which are similar to [20], the news between 2006-10-01 and 2012-12-31 for training, had news from 2013-01-01 to 2013-06-15 for validating and testing contained news from 16-06-2013 to 31-12-2013.”
Lancaster A at SemEval-2017 Task 5: Evaluation metrics matter: predicting sentiment from financial news headlines (2017, background): “, 2016) to stock market prediction models (Peng and Jiang, 2016; Kazemian et al., 2016).”
Applications of deep learning in stock market prediction: recent progress (2020, background): “Li et al. (2019b) formulates a sentiment-ARMA model to incorporate the news articles as hidden information and designs a LSTM-based DNN, which consists of three components, namely, LSTM, VADER model and differential privacy mechanism that integrates different news sources.”
Knowledge-Driven Event Embedding for Stock Prediction (2016, background): “Stock Market Prediction There has been a line of work predicting stock markets using text information from daily news (Lavrenko et al., 2000; Schumaker and Chen, 2009; Xie et al., 2013; Peng and Jiang, 2015; Li et al., 2016).”