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
Sentiment analysis of Swedish social media: Using random indexing to improve cross-domain sentiment classification
KTH, School of Computer Science and Communication (CSC).
2014 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

 Social media has grown extremely fast in recent years and in the vast number of posts being made everyday people express their opinions about all kinds of topics. These opinions are very valuable and there is a need for a way to automatically identify and extract them. This is what sentiment analysis is about but there are a number of issues related to this task. In particular the large number and diversity of the texts to analyze causes problems for ordinary methods of natural language processing. In this thesis a method utilizing a technique called Random Indexing is proposed which tries to overcome some of the issues. The conclusion is that the use of Random Indexing does aid in solving the problem but also that more work is needed in order to have a fully satisfying solution.

Abstract [sv]

Sentimentanalys av svenska sociala medier

Användningen av sociala medier har vuxit snabbt de senaste åren och i den stora mängd inlägg som skrivs varje dag gömmer sig många människors åsikter. Dessa åsikter innehåller värdefull information och det behövs ett sätt att automatiskt identifiera och ta tillvara på den. Sentimentanalys behandlar precis detta men det finns ett antal svårigheter med att lösa denna uppgift. Svårigheterna rör framförallt att det finns en så stor mängd texter att analysera och hur väldigt olika de kan vara. I det här exjobbet föreslås en metod som använder sig av en teknik kallad Random Indexing för att överkomma vissa av dessa svårigheter. Slutsatsen är att användningen av Random Indexing hjälper till att lösa problemen men att det fortfarande krävs mer arbete för att få fram en fullt fungerande lösning.

Place, publisher, year, edition, pages
2014.
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-153682OAI: oai:DiVA.org:kth-153682DiVA: diva2:753301
Examiners
Available from: 2015-05-28 Created: 2014-10-07 Last updated: 2015-05-28Bibliographically approved

Open Access in DiVA

fulltext(585 kB)440 downloads
File information
File name FULLTEXT01.pdfFile size 585 kBChecksum SHA-512
092760f10085100e2944b5a186e0aa5d76065c26f311aa5542889351c3315c4d5098939c6771d543140fc0cc4cef9ffb2c52cb2b201dd40f3ad03bf7f3df12f4
Type fulltextMimetype application/pdf

By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 440 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 110 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