Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Machine Learning News Analysis for Stock Trading.
KTH, School of Computer Science and Communication (CSC).
2012 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

This thesis proposes an autonomous module for news article analysis. The module is training a classifier on historical news articles where the market reaction is known. Using the trained classifier on freshly published news articles, the module tries to predict whether the article is good or bad and whether the stock of the concerned company will go up or down in the near future (up to a week).

Prediction of up or down better than chance was beaten in theory--its practical usage is limited though. However, on the other point of evaluation (good or bad article) the module performed very well.

The forecast was aggregated with a forecast from another module, which was based on price pattern analysis. The aggregation is discussed in this report but the price pattern analysis module is not.

Abstract [sv]

Denna rapport föreslår en autonom modul för nyhetsanalys. Modulen tränar en klassificerare på historiska nyhetsartiklar där marknadsreaktionen är känd. När sedan den tränade klassificeraren används på precis publicerade nyheter försöker modulen förutspå ifall artikeln var bra eller dålig och ifall berört företags aktie kommer gå upp eller ner i närtid (upp till en vecka).

Att kunna förutspå upp eller ner bättre än chansen evaluerades och klarades av i teorin men blir svår att nyttja i parktiken. Däremot att kunna avgöra ifall en artikel var bra eller dålig presterade modulen väldigt bra på.

Förutsägelsen aggregerades även med en förutsägelse från en annan modul. Denna var baserad på prismönsteranalys. Aggregeringen diskuteras i rapporten men inte prismönsteranalysmodulen.

Place, publisher, year, edition, pages
2012.
Series
Trita-CSC-E, ISSN 1653-5715 ; 2012:089
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-130934OAI: oai:DiVA.org:kth-130934DiVA: diva2:654380
Educational program
Master of Science in Engineering - Computer Science and Technology
Uppsok
Technology
Supervisors
Examiners
Available from: 2013-10-07 Created: 2013-10-07

Open Access in DiVA

No full text

Other links

http://www.nada.kth.se/utbildning/grukth/exjobb/rapportlistor/2012/rapporter12/elgh_johannes_12089.pdf
By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 119 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf