Analysis of news sentiments using natural language processing and deep learning
- PDF / 680,602 Bytes
- 7 Pages / 595.276 x 790.866 pts Page_size
- 63 Downloads / 192 Views
OPEN FORUM
Analysis of news sentiments using natural language processing and deep learning Mattia Vicari1 · Mauro Gaspari2 Received: 29 October 2020 / Accepted: 2 November 2020 © The Author(s) 2020
Abstract This paper investigates if and to what point it is possible to trade on news sentiment and if deep learning (DL), given the current hype on the topic, would be a good tool to do so. DL is built explicitly for dealing with significant amounts of data and performing complex tasks where automatic learning is a necessity. Thanks to its promise to detect complex patterns in a dataset, it may be appealing to those investors that are looking to improve their trading process. Moreover, DL and specifically LSTM seem a good pick from a linguistic perspective too, given its ability to “remember” previous words in a sentence. After having explained how DL models are built, we will use this tool for forecasting the market sentiment using news headlines. The prediction is based on the Dow Jones industrial average by analyzing 25 daily news headlines available between 2008 and 2016, which will then be extended up to 2020. The result will be the indicator used for developing an algorithmic trading strategy. The analysis will be performed on two specific cases that will be pursued over five time-steps and the testing will be developed in real-world scenarios. Keywords Deep learning · Machine learning · Natural language processing · Trading signals · Trading · Sentiment analysis · NLP · Trading strategies
1 Introduction Stock forecasting through NLP is at the crossroad between linguistics, machine learning, and behavioral finance (Xing et al. 2018). One of the main NLP techniques applied on financial forecasting is sentiment analysis (Cambria 2016) which concerns the interpretation and classification of emotions within different sources of text data. It is a research area revived in the last decade due to the rise of social media and cheap computing power availability (Brown 2016). Like products and services, market sentiments influence information flow and trading, thus trading firms hope to profit based on forecasts of price trends influenced by sentiments in financial news (Ruiz-Martínez et al. 2012). Is it possible
* Mattia Vicari [email protected] Mauro Gaspari [email protected] 1
University of Bologna, Bologna, Italy
Department of Computer Science and Engineering, University of Bologna, Bologna, Italy
2
to find predictive power in the stock market’s behavior based on them? It seems to be the case in the work “On the importance of text analysis for stock market prediction” by Lee and MacCartney (2014) that shows, based on text, an improved predictability in the performance of a security. Intuitively, the cause of the stocks’ fluctuation can be the aggregated behavior of the stockholders, who will act based on news (Xing et al. 2018). Although the predicting models reported in the literature have not been able to profit in the long run, many theories and meaningful remarks have been made from t
Data Loading...
