Abstract
The explosive growth of online news and the need to find the right news article quickly and efficiently cause people to adapt on events happening. The readers task is to filter out the desired information from headlines and teasers by scanning various sources formats (text, broadcasting transmission and video storage) of news articles. What people need are entities, relationships and events, which can be extracted from text by using event extraction techniques. Considering the granularity of event extraction, we present a novel approach that extracts correlation between news with a human/machine interaction. Our scope is to answer this question more efficiently: “How might stocks (e.g. Eni) react if a news is created and launched again across web news network?”. This research examines a predictive machine learning approach for financial news articles analysis using a News Index Map (NIM) based on a web news decision support system for event forecasting and trading decision. Empirical evaluation on real online news data sets firstly show that only a small number of news ends up having a real impact on the security and secondly, human coding is able to extract knowledge from large amounts of data to build predictive models to provide investment decision suggestions.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
NewsMarket is a prototype created by QBT Sagl.
- 2.
Eni is an integrated energy company. Active in more than 70 countries, with a staff of 76,000 employees, in the oil and gas, electricity generation and sale, petrochemicals, oilfield services construction and engineering industries.
- 3.
Xie, Hari, and Campbell [11] write: “Events can be defined as real-world occurrences that unfold over space and time. In other words, an event has a duration, occurs in a specific place, and typically will involve certain change of state. Using this definition, “a walk on the beach”, “the hurricane of 2005”, and “a trip to Santa Barbara” would all qualify as events. Events are useful because they help us make sense of the world around us by helping to recollect real-world experiences (e.g., university commencement 2006), by explaining phenomena that we observe (e.g., the annual journey of migrating birds), or by assisting us in predicting future events (e.g., the outcome of a tennis match).”
- 4.
In this study we have excluded audio and video news. We will consider them for the development of the definite prototype.
- 5.
Data was taken from Bloomberg, il sole 24 ore website, borsainside.com, Trend-online.com and Eni.it.
- 6.
Although the trading starts at 9:05 am and close at 17:30 CET we felt important to consider news article release during all the day (also news posted after closing hours) and for every days (during weekends, or on holidays).
- 7.
The FTSE MIB Index is the primary benchmark index for the Italian equity market.
References
Casti J (2012) Eventi X, Eventi estremi e il futuro della civilta’. Il Saggiatore
Grimmer J, Stewart BM (2013) Text as data: The Promise and Pitfalls of automatic content analysis methods for political texts. Pol Anal 21(3):267–297
Haas LM, Kossmann D, Wimmers EL, Yang J (1997) Optimizing queries across diverse data sources. In: Proceedings of the twenty-third international conference on very large data bases, Athens, pp 276–285
Jonathan C, Boyd-Graber J, Wang C, Gerrish S, Blei DM (2009) Reading tea leaves: how humans interpret topic models. In: Bengio Y, Schuurmans D, Lafferty J, Williams CKI, Culotta A (eds) Advances in neural information processing systems. MIT, Cambridge, pp 288–296
Kloptchenko A, Eklund T et al (2004) Combining data and text mining techniques for analysing financial reports. Intell Syst Acc Financ Manag 12(1):29–41
Mittermayer M (2004) Forecasting intraday stock price trends with text mining techniques. In: Hawaii international conference on system sciences, Kailua-Kona
Scheffer M (2009) Critical transitions in nature and society. Princeton studies in complexity. Princeton University Press, Princeton
Schumaker RP, Chen H (2006) Textual analysis of stock market prediction using financial news articles. In: 12th Americas conference on information systems (AMCIS-2006), Acapulco
Tsoi AC, Tan CNW, Lawrence S (1993) Financial time series forecasting: application of artificial neural network techniques. In: 1993 international symposium on nonlinear theory and its applications, Hawaii
Ullman JD (1997) Information integration using logical views. In: Proceedings of the sixth international conference on database theory, Delphi. pp 19–40
Xie L, Sundaram H, Campbell M (2008) Event mining in multimedia streams. Proc IEEE 96(4):623–647
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Barazzetti, A., Cecconi, F., Mastronardi, R. (2014). Financial Forecasts Based on Analysis of Textual News Sources: Some Empirical Evidence. In: Leitner, S., Wall, F. (eds) Artificial Economics and Self Organization. Lecture Notes in Economics and Mathematical Systems, vol 669. Springer, Cham. https://doi.org/10.1007/978-3-319-00912-4_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-00912-4_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-00911-7
Online ISBN: 978-3-319-00912-4
eBook Packages: Business and EconomicsEconomics and Finance (R0)