Show simple item record

dc.contributor.authorKvifte, Tord
dc.date.accessioned2021-06-21T07:34:24Z
dc.date.issued2021-06-01
dc.date.submitted2021-06-18T22:00:21Z
dc.identifier.urihttps://hdl.handle.net/11250/2760300
dc.descriptionPostponed access: the file will be accessible after 2022-06-01
dc.description.abstractWhen a movie is uploaded to a movie Recommender System (e.g., YouTube), the system can exploit various forms of descriptive features (e.g., tags and genre) in order to generate personalized recommendation for users. However, there are situations where the descriptive features are missing or very limited and the system may fail to include such a movie in the recommendation list, known as Cold-start problem. This thesis investigates recommendation based on a novel form of content features, extracted from movies, in order to generate recommendation for users. Such features represent the visual aspects of movies, based on Deep Learning models, and hence, do not require any human annotation when extracted. The proposed technique has been evaluated in both offline and online evaluations using a large dataset of movies. The online evaluation has been carried out in a evaluation framework developed for this thesis. Results from the offline and online evaluation (N=150) show that automatically extracted visual features can mitigate the cold-start problem by generating recommendation with a superior quality compared to different baselines, including recommendation based on human-annotated features. The results also point to subtitles as a high-quality future source of automatically extracted features. The visual feature dataset, named DeepCineProp13K and the subtitle dataset, CineSub3K, as well as the proposed evaluation framework are all made openly available online in a designated Github repository.
dc.language.isoeng
dc.publisherThe University of Bergen
dc.rightsCopyright the Author. All rights reserved
dc.subjectuser-centric evaluation
dc.subjecthybrid recommender systems
dc.subjectimage recognition
dc.subjectdeep learning
dc.subjectrecommender systems
dc.subjectmachine learning
dc.titleVideo Recommendations Based on Visual Features Extracted with Deep Learning
dc.typeMaster thesis
dc.date.updated2021-06-18T22:00:21Z
dc.rights.holderCopyright the Author. All rights reserved
dc.description.degreeMasteroppgave i informasjonsvitenskap
dc.description.localcodeINFO390
dc.description.localcodeMASV-INFO
dc.subject.nus735115
fs.subjectcodeINFO390
fs.unitcode15-17-0
dc.date.embargoenddate2022-06-01


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record