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
Tweetopolitics: A Scalable Platform for Analyzing Swedish Elections on Twitter
KTH, School of Information and Communication Technology (ICT).
2015 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Due to its real-time, open and democratic nature in information dissemination Twitter has become a common-ground for civic opinion expression as well as observation. This is much anticipated on the wake of an electoral event. As a result, politicians spend considerable amount of resources on analyzing the positive and negative opinions shared with respect to their orientation. Ultimately they become keen on leveraging their popularity to increase their followers base through social media. This is where user profiling and recommendation systems become useful by eliciting social preferences of political parties and their respective followers, and leveraging such intelligence to suggest relevant parties to the interested followers. To this end, we present results of applying our user profiling approach on a Twitter dataset focus-crawled during the course of 2014 general elections of Sweden. Results of analyzing content of tweets show how parties speak of controversial topics, while results of network mining show how much interactions and popularities of the parties change during the month of Election.

Abstract [sv]

På grund av dess realtid, har öppet och demokratiskt naturen i informationsspridning Twitter blivit en vanlig mark för medborgar yttrande uttryck samt observation. Detta är mycket väntat på spåren av en valhändelse. Som ett resultat, politiker spenderar betydande resurser på att analysera de positiva och negativa yttranden delas med avseende på deras orientering. Ytterst blir de angelägna om att utnyttja sin popularitet att öka sin efterföljare bas via sociala medier. Det är där användarprofilering och rekommendationssystem blir användbar genom att framkalla sociala preferenser för politiska partier och deras respektive anhängare, och utnyttja sådan intelligens att föreslå relevanta parter till de intresserade anhängare. Därför presenterar vi resultatet av att tillämpa våra användarprofilering riktlinje om ett Twitter dataset fokus kröp under 2014 allmänna val i Sverige. Resultat av analys innehåll tweets visa hur partierna talar om kontroversiella ämnen, medan resultatet av nätverksgruv visa hur mycket interaktion och popularities parternas förändras under månaden valet.

Place, publisher, year, edition, pages
2015. , 80 p.
Series
TRITA-ICT-EX, 2015:257
Keyword [en]
twitter, political parties profiling, stagger, lda, wordle, twitter crawler, mongodb, user recommendation
Keyword [sv]
twitter, political parties profiling, stagger, lda, wordle, twitter crawler, mongodb, user recommendation
National Category
Software Engineering
Identifiers
URN: urn:nbn:se:kth:diva-210822OAI: oai:DiVA.org:kth-210822DiVA: diva2:1120313
Subject / course
Information and Software Systems
Educational program
Master of Science - Software Engineering of Distributed Systems
Examiners
Available from: 2017-07-06 Created: 2017-07-06 Last updated: 2017-07-06Bibliographically approved

Open Access in DiVA

No full text

By organisation
School of Information and Communication Technology (ICT)
Software Engineering

Search outside of DiVA

GoogleGoogle Scholar

Total: 1 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