Automated news reading: Stock price prediction based on financial news using context-capturing features

We examine whether stock price prediction based on textual information in financial news can be improved as previous approaches only yield prediction accuracies close to guessing probability. Accordingly, we enhance existing text mining methods by using more expressive features to represent text and...

Full description

Saved in:
Bibliographic Details
Published in:Decision Support Systems Vol. 55; no. 3; pp. 685 - 697
Main Authors: Hagenau, Michael, Liebmann, Michael, Neumann, Dirk
Format: Journal Article
Language:English
Published: Amsterdam Elsevier B.V 01.06.2013
Elsevier
Elsevier Sequoia S.A
Subjects:
Applied sciences
Artificial intelligence
Classification
Computer science; control theory; systems
Data entry
Data mining
Decision support
Decision support systems
Exact sciences and technology
Feature selection
Financial news
Hoisting
Information systems. Data bases
Inventory control, production control. Distribution
Machine learning
Markets
Memory organisation. Data processing
News
News media
Operational research and scientific management
Operational research. Management science
Portfolio theory
Probability
Raw materials
Software
Speech and sound recognition and synthesis. Linguistics
Stock price prediction
Stock prices
Studies
Text mining
Texts
Text mining
Financial news
Stock price prediction
Decision support
Textual data
Probabilistic approach
Process selection
Feedback regulation
Financial market
Text
Data mining
Production management
Feedback
Stock exchange
Semantics
Learning (artificial intelligence)
Classification
Selection criterion
Financial data
Automatic recognition
Artificial intelligence
Reading device
ISSN:0167-9236, 1873-5797
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We examine whether stock price prediction based on textual information in financial news can be improved as previous approaches only yield prediction accuracies close to guessing probability. Accordingly, we enhance existing text mining methods by using more expressive features to represent text and by employing market feedback as part of our feature selection process. We show that a robust feature selection allows lifting classification accuracies significantly above previous approaches when combined with complex feature types. This is because our approach allows selecting semantically relevant features and thus, reduces the problem of over-fitting when applying a machine learning approach. We also demonstrate that our approach is highly profitable for trading in practice. The methodology can be transferred to any other application area providing textual information and corresponding effect data. ► We introduce a 2-word text representation scheme for better capturing of context and meaning. ► Feature selection employs market feedback to pick the most relevant features. ► We lift classification accuracies significantly above previous approaches. ► Trading simulation shows that our method can be applied in practice and is profitable. ► Method can be transferred to any other application area providing text and feedback.
Bibliography:SourceType-Scholarly Journals-1
content type line 14
ObjectType-Editorial-2
ObjectType-Commentary-1
ObjectType-Article-2
ObjectType-Feature-1
content type line 23
ISSN:0167-9236
1873-5797
DOI:10.1016/j.dss.2013.02.006