Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Ontology-based personalised retrieval in support of reminiscence

Shi, Lei and Setchi, Rossitza ORCID: 2013. Ontology-based personalised retrieval in support of reminiscence. Knowledge-Based Systems 45 , pp. 47-61. 10.1016/j.knosys.2013.02.004

Full text not available from this repository.


This research proposes a knowledge-based framework for integrating ontology-based personalised retrieval and reminiscence support. The aim is to assist people in recalling, browsing and re-discovering events from their lives by considering their profiles and background knowledge and providing them with customised information retrieval. To model a user’s background knowledge, this paper defines a user profile space (UPS) model and describes its construction method. The model has a dynamic structure based on relevance feedback and interactions with users. Furthermore, this work introduces a multi-ontology query expansion model which uses user-oriented ontologies, UPSs and semantic feature-selection algorithms to expand queries. In this model, knowledge-spanning trees are generated from ontology/UPS graphs based on the queries. These knowledge-spanning trees contain semantic features which enhance the representations of the original queries and further facilitate personalised retrieval on a semantic basis. The experimental results indicate that the proposed approach consistently outperforms term-based retrieval on precision, recall and f-score, which proves the positive effect of using ontology/user profile spaces in query expansion and personalised retrieval.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Engineering
Subjects: Q Science > QA Mathematics > QA76 Computer software
T Technology > T Technology (General)
Uncontrolled Keywords: Ontology graph; User profile space; Knowledge spanning tree; Feature selection; Personalised retrieval; Reminiscence support
Publisher: Elsevier
ISSN: 0950-7051
Last Modified: 06 Jul 2023 10:18

Citation Data

Cited 15 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item Edit Item