Abstract
A simple and novel method for generating labeled examples for sentiment analysis is introduced: news stories about publicly traded companies are labeled positive or negative according to price changes of the company stock. It is shown that there are many lexical markers for bad news but none for good news. Overall, learned models based on lexical features can distinguish good news from bad news with accuracy of about 70%. Unfortunately, this result does not yield profits since it works only when stories are labeled according to cotemporaneous price changes but does not work when they are labeled according to subsequent price changes.
Keywords
Yield (engineering)News analyticsStock (firearms)Stock priceNews mediaSentiment analysisComputer scienceStock marketSimple (philosophy)BusinessAdvertisingEconomicsEconometricsArtificial intelligenceHistoryMachine learning
Related Publications
Liquidity and the 1987 stock market crash
T he Crash of October 1987 was a puzzling event puzzling not only for what happened on October 19, but also for what subsequently did not happen. In spite of the magnitude and m...
Data dredging, bias, or confounding
On 4 October 2002, women who were moderate drinkers received good news: their risk of breast cancer was not raised, according to a report in the Lancet that was widely covered b...
SECURITY PRICES, RISK, AND MAXIMAL GAINS FROM DIVERSIFICATION*
Sections II and III of this paper set forth the simple logic which leads directly to the determination of explicit equilibrium prices of risk assets traded in competitive market...
How Capital Budgeting Deters Innovation—And What To Do About It
When a company fails to stay dominant in key markets, internal systems of resource allocation may be the cause. decline often begins when financial managers impose strict contr...
The theory of the parent–infant relationship
The main point of this paper can perhaps best be brought out through a comparison of the study of infancy with the study of the psycho-analytic transference.2 It cannot be too s...
Publication Info
- Year
- 2004
- Type
- article
- Citations
- 51
- Access
- Closed
External Links
Citation Metrics
51
OpenAlex
Cite This
Moshe Koppel,
Itai Shtrimberg
(2004).
Good News or Bad News? Let the Market Decide.
National Conference on Artificial Intelligence
.