Hybrid Reciprocal Recommender Systems: Integrating Item-to-User Principles in Reciprocal Recommendation

James Neve, Iván Palomares

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

Reciprocal Recommender Systems (RRS) recommend users to other users in a personalised manner, in scenarios where both sides of the preference relation must be considered. Existing RRS approaches based on collaborative filtering or content-based filtering, have been used for enhancing user experience in online dating and other online services aimed at connecting users with each other. However, some of these services e.g. skill sharing platforms, are still pervaded by content published, shared and consumed by users, consequently there is a valuable source of item-to-user preferential information not captured by existing RRS models. We present a novel hybrid RRS framework that integrates user preferences towards content in reciprocal recommendation, and we instantiate and evaluate it using data from Cookpad, a recipe sharing social media platform. As part of our model, we also implement a novel content-based extension of Jaccard similarity measure that operates on word embeddings.

Original languageEnglish
Title of host publicationThe Web Conference 2020 - Companion of the World Wide Web Conference, WWW 2020
PublisherAssociation for Computing Machinery
Pages848-853
Number of pages6
ISBN (Electronic)9781450370240
DOIs
Publication statusPublished - 2020 Apr 20
Event29th International World Wide Web Conference, WWW 2020 - Taipei, Taiwan
Duration: 2020 Apr 202020 Apr 24

Publication series

NameThe Web Conference 2020 - Companion of the World Wide Web Conference, WWW 2020

Conference

Conference29th International World Wide Web Conference, WWW 2020
Country/TerritoryTaiwan
CityTaipei
Period20-04-2020-04-24

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Software

Fingerprint

Dive into the research topics of 'Hybrid Reciprocal Recommender Systems: Integrating Item-to-User Principles in Reciprocal Recommendation'. Together they form a unique fingerprint.

Cite this