Zur Seitenansicht


Movie genome: alleviating new item cold start in movie recommendation
AutorInnenSchedl, Markus ; Deldjoo, Yashar ; Dacrema, Maurizio Ferrari ; Constantin, Mihai Gabriel ; Eghbal-zadeh, Hamid ; Cereda, Stefano ; Ionescu, Bogdan ; Cremonesi, Paole
Erschienen in
User Modeling and User-Adapted Interaction, 2019,
ErschienenSpringer Netherlands, 2019
DokumenttypAufsatz in einer Zeitschrift
Schlagwörter (EN)Movie recommender systems / Cold start / Warm start / Semi-cold start / New item / Multimedia features / Content-based / Audio descriptors / Visual descriptors / Multimodal fusion / Hybrid recommender system / Feature weighting / Collaborative-enriched content-based filtering / Canonical correlations analysis
URNurn:nbn:at:at-ubl:3-1901 Persistent Identifier (URN)
 Das Werk ist gemäß den "Hinweisen für BenützerInnen" verfügbar
Movie genome: alleviating new item cold start in movie recommendation [1.41 mb]
Zusammenfassung (Englisch)

As of today, most movie recommendation services base their recommendations on collaborative filtering (CF) and/or content-based filtering (CBF) models that use metadata (e.g., genre or cast). In most video-on-demand and streaming services, however, new movies and TV series are continuously added. CF models are unable to make predictions in such a scenario, since the newly added videos lack interactionsa problem technically known as new item cold start (CS). Currently, the most common approach to this problem is to switch to a purely CBF method, usually by exploiting textual metadata. This approach is known to have lower accuracy than CF because it ignores useful collaborative information and relies on human-generated textual metadata, which are expensive to collect and often prone to errors. User-generated content, such as tags, can also be rare or absent in CS situations. In this paper, we introduce a new movie recommender system that addresses the new item problem in the movie domain by (i) integrating state-of-the-art audio and visual descriptors, which can be automatically extracted from video content and constitute what we call the movie genome; (ii) exploiting an effective data fusion method named canonical correlation analysis, which was successfully tested in our previous works Deldjoo et al. (in: International Conference on Electronic Commerce and Web Technologies. Springer, Berlin, pp 3445, 2016b; Proceedings of the Twelfth ACM Conference on Recommender Systems. ACM, 2018b), to better exploit complementary information between different modalities; (iii) proposing a two-step hybrid approach which trains a CF model on warm items (items with interactions) and leverages the learned model on the movie genome to recommend cold items (items without interactions). Experimental validation is carried out using a system-centric study on a large-scale, real-world movie recommendation dataset both in an absolute cold start and in a cold to warm transition; and a user-centric online experiment measuring different subjective aspects, such as satisfaction and diversity. Results show the benefits of this approach compared to existing approaches.

Das PDF-Dokument wurde 3 mal heruntergeladen.
CC-BY-Lizenz (4.0)Creative Commons Namensnennung 4.0 International Lizenz