Titlebar

Export bibliographic data
Literature by the same author
plus on the publication server
plus at Google Scholar

 

Supporting Serendipitous Recommendations With Knowledge Graphs

Title data

Baumann, Oliver ; Schönfeld, Mirco:
Supporting Serendipitous Recommendations With Knowledge Graphs.
2022
Event: 2nd Joint Conference of the Information Retrieval Communities in Europe (CIRCLE 2022) , 04.-07. Juli 2022 , Samatan, Gers, France.
(Conference item: Conference , Paper )

Official URL: Volltext

Related URLs

Project information

Project title:
Project's official titleProject's id
Africa Multiple Cluster of Excellence at the University of BayreuthEXC 2052/1 – 390713894

Project financing: Deutsche Forschungsgemeinschaft

Abstract in another language

Recommender systems are commonly designed and evaluated with high precision and accuracy in mind. Optimising systems for these metrics alone can, however, lead to a decrease in overall collection coverage of recommended items, and carries potential to over-emphasize popular content. Notions such as serendipitous discovery and, closely related, novelty and diversity propose that rather than replicating a user's taste or that of a cluster of others they happen to be similar to, recommendation systems should present useful suggestions to users, including novel and diverse items and thus supporting serendipitous discovery. We implement a recommender system based on a knowledge graph of musical items with serendipity, novelty and diversity in mind. Using acoustic features as contextual information for vertices in the graph, we explicitly select content dissimilar from the user's previous experience. We compare our results to a set of baseline algorithms and find that we are able to recommend diverse and novel items.

Further data

Item Type: Conference item (Paper)
Refereed: Yes
Keywords: information retrieval; knowledge graphs; recommender systems; serendipity; novelty; diversity; attributed graphs
Institutions of the University: Faculties > Faculty of Languages and Literature
Faculties > Faculty of Languages and Literature > Juniorprofessur Datenmodellierung und interdisziplinäre Wissensgenerierung
Faculties > Faculty of Languages and Literature > Juniorprofessur Datenmodellierung und interdisziplinäre Wissensgenerierung > Juniorprofessur Datenmodellierung und interdisziplinäre Wissensgenerierung - Juniorprof. Dr. Mirco Schönfeld
Faculties
Result of work at the UBT: Yes
DDC Subjects: 000 Computer Science, information, general works > 004 Computer science
Date Deposited: 02 Aug 2022 06:36
Last Modified: 05 Aug 2022 08:55
URI: https://eref.uni-bayreuth.de/id/eprint/70526