Picture of classic books on shelf

Literary linguistics: Open Access research in English language

Strathprints makes available Open Access scholarly outputs by English Studies at Strathclyde. Particular research specialisms include literary linguistics, the study of literary texts using techniques drawn from linguistics and cognitive science.

The team also demonstrates research expertise in Renaissance studies, researching Renaissance literature, the history of ideas and language and cultural history. English hosts the Centre for Literature, Culture & Place which explores literature and its relationships with geography, space, landscape, travel, architecture, and the environment.

Explore all Strathclyde Open Access research...

Popularity over Relevance in Collaborative Tagging Systems for General Resource Discovery

Macgregor, George and McCulloch, Emma (2005) Popularity over Relevance in Collaborative Tagging Systems for General Resource Discovery. Discussion paper. University of Strathclyde, Glasgow. (Unpublished)

[img]
Preview
Text (Macgregor-McCulloch-POR-2005-Popularity-over-relevance-in-collaborative-tagging-systems)
Macgregor_McCulloch_POR_2005_Popularity_over_relevance_in_collaborative_tagging_systems.pdf
License: Creative Commons Attribution 4.0 logo

Download (530kB) | Preview

Abstract

Although services offering collaborative tagging provide rich interactive experiences that exemplify a sense of community among disparate user groups, their raison d'être ultimately remains similar to traditional information retrieval systems. In this paper we will explore potential conflict between user requirements for PIM and CIM, and the theme of popularity over relevance by revisiting basic information retrieval principles within the context of collaborative tagging. The long-term implications of these factors for the efficacy of collaborative tagging in general resource discovery will also be examined. The theoretical analysis presented suggests that PIM requirements conflict with the wider requirements of resource discovery by invoking a series of extra relevance variables and that emphasis on popularly used tags and the subsequent effect on ‘collective behavior’ potentially compromises the effectiveness of system performance. Hypotheses are proposed for testing and further research areas are suggested.