Change search
ReferencesLink to record
Permanent link

Direct link
Extending recommendation algorithms bymodeling user context
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]

Recommender systems have been widely adopted by onlinee-commerce websites like Amazon and music streaming services like Spotify. However, most research efforts have not sufficiently considered the context in which recommendations are made, especially when the input is implicit.In this work, we investigate the value of including contextual information like day-of-week in collaborative filtering recommender systems. For the investigation, we first implemented two algorithms, namely contextual prefiltering and contextual post-filtering. Then, we evaluated these algorithms with user data collected from Spotify.Experiment results show that the pre-filtering algorithm shows some promise against an item similarity baseline, indicating that further investigation could be rewarding. The post-filtering algorithm underperforms a popularity-derived baseline, due to information loss in the recommendation process.

Abstract [sv]

Förbättrade rekommenadtionsalgoritmergenom att använda användarens kontext. Rekommendationssystem har spridda användsningområden så som e-handels företag som Amazon och internetbaserade musiktjänster som Spotify. Mesta forskningen inom rekommendationssystem har inte tagit användares context i beaktning och speciellt inte då datan är av implicit typ. I det här projektet har vi undersökt vikten av att inkludera information om användares context, så som veckodag, itraditionella rekommendationssystem baserat på collaborative filtering. Vi har implementerat två algoritmer, contextualpre-filtering och contextual post-filtering och utvärderatdem på användardata från Spotify. Experimenten visar att post-filtering algoritmen presterarsämre än en standard popularitetsbaserad algoritm på grund utav informationsförlust i rekommendationssteget. Algoritmen baserad på pre-filtering visar lovande resultatjämfört med en standard item similarity algoritm vilket lovordar vidare undersökningar.

Place, publisher, year, edition, pages
National Category
Computer Science
URN: urn:nbn:se:kth:diva-154044OAI: diva2:754842
Available from: 2014-11-21 Created: 2014-10-13 Last updated: 2014-11-21Bibliographically approved

Open Access in DiVA

fulltext(770 kB)196 downloads
File information
File name FULLTEXT01.pdfFile size 770 kBChecksum SHA-512
Type fulltextMimetype application/pdf

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

Search outside of DiVA

GoogleGoogle Scholar
Total: 196 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: 83 hits
ReferencesLink to record
Permanent link

Direct link