Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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
Modelling an individual's selection of a partner in a speed-dating experiment using a priori knowledge
KTH, School of Computer Science and Communication (CSC).
2017 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Modellera en individs val av partner i ett speed-dating experiment med a priori kunskap (Swedish)
Abstract [en]

Speed dating is a relative new concept that allows researchers to study various theories related to mate selection. A problem with current research is that it focuses on finding general trends and relationships between the attributes.

This report explores the use of machine learning techniques to predict whether an individual will want to meet his partner again after the 4-minute meeting based on their attributes that were known before they met. We will examine whether Random Forest or Extremely Randomized Trees perform better than Support Vector Machines for both limited attributes (describe appearance only) and extended attributes (includes answers to some questions about their preferences).

It is shown that Random Forests perform better than Support Vector Machines and that extended attributes give better result for both classifiers. Furthermore, it is observed that the more information is known about the individuals, the better a classifier performs. Clubbing preferences of the partner stands out as an important attribute, followed by the same preference for the individual.

Abstract [sv]

Speed dating är ett relativt nytt koncept som tillåter forskare att studera olika teorier relaterade till val av partner. Ett problem med nuvarande forskning är att den fokuserar på att hitta generella trender och samband mellan attribut.

Den här rapporten utforskar användning av maskinlärningsteknik för att förutsäga om en individ kommer vilja träffa sin partner igen efter ett 4-minuters möte baserat på deras attribut som var tillgängliga innan de träffades. Vi kommer att undersöka om Random Forest eller Extremely Randomized Trees fungerar bättre än Support Vector Machine för både begränsade attribut (beskriver bara utseende) och utökade attribut (inkluderar svar på några frågor om deras preferenser).

Det visas att Random Forest fungerar bättre än Support Vector Machines och att utökade attribut ger bättre resultat för båda klassificerarna. Dessutom är det observerat att ju mer information som finns tillgänglig om individerna, desto bättre resultat ger en klassificerare. Partners preferens för att besöka nattklubbar står ut som ett viktigt attribut, följt av individers samma preferens för individen.

Place, publisher, year, edition, pages
2017.
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-208668OAI: oai:DiVA.org:kth-208668DiVA: diva2:1107797
Supervisors
Examiners
Available from: 2017-06-18 Created: 2017-06-10 Last updated: 2017-06-18Bibliographically approved

Open Access in DiVA

fulltext(679 kB)33 downloads
File information
File name FULLTEXT01.pdfFile size 679 kBChecksum SHA-512
df52a1d689b71037075d432ad29e0e528aa4ae17b18ce98279580a1c7488bca922566b0d9bcddd78cdf685414943281d4ebd0266fc44517ae518a48d1dc285ee
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Los, Artem
By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 33 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

Total: 135 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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