Login (DCU Staff Only)
Login (DCU Staff Only)

DORAS | DCU Research Repository

Explore open access research and scholarly works from DCU

Advanced Search

Social contextuality and conversational recommender systems

Hurrell, Eoin (2013) Social contextuality and conversational recommender systems. PhD thesis, Dublin City University.

Abstract
As people continue to become more involved in both creating and consuming information, new interactive methods of retrieval are being developed. In this thesis we examine conversational approaches to recommendation, that is, the act of suggesting items to users based on the system’s understanding of them. Conversational recommendation is a recent contribution to the task of information discovery. We propose a novel approach to conversation around recommendation, examining how it is improved to work with collaborative filtering, a common recommendation algorithm. In developing new ways to recommend information to people we also examine their methods of information seeking, exploring the role of conversational recommendation, using both interview and sensed brain signals. We also look at the implications of the wealth of social and sensed information now available and how it improves the task of accurate recommendation. By allowing systems to better understand the connections between users and how their social impact can be tracked we show improved recommendation accuracy. We look at the social information around recommendations, proposing a directed influence approach between socially connected individuals, for the purpose of weighting recommendations with the wisdom of influencers. We then look at the semantic relationships that might seem to indicate wisdom (i.e. authors on a book-ranking site) to see if the ``wisdom of the few'' can be traced back to those conventionally considered wise in the area. Finally we look at ``contextuality'' (the ability of sets of contextual sensors to accurately recommend items across groups of people) in recommendation, showing that different users have very different uses for context within recommendation. This thesis shows that conversational recommendation can be generalised to work well with collaborative filtering, that social influence contributes to recommendation accuracy, and that contextual factors should not be treated the same for each user.
Metadata
Item Type:Thesis (PhD)
Date of Award:March 2013
Refereed:No
Supervisor(s):Smeaton, Alan F.
Uncontrolled Keywords:Recommender Systems; Brain Computer Interfaces
Subjects:Computer Science > Computer software
Computer Science > Multimedia systems
DCU Faculties and Centres:DCU Faculties and Schools > Faculty of Engineering and Computing > School of Computing
Research Initiatives and Centres > CLARITY: The Centre for Sensor Web Technologies
Use License:This item is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 3.0 License. View License
Funders:Science Foundation Ireland
ID Code:17737
Deposited On:03 Apr 2013 12:46 by Alan Smeaton . Last Modified 19 Jul 2018 14:58
Documents

Full text available as:

[thumbnail of thesis_final.pdf]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
7MB
Downloads

Downloads

Downloads per month over past year

Archive Staff Only: edit this record