Caroline Williams, Executive Director of Intute [email protected] Andy Priest, Intute Technical Co- ordinator [email protected] PERsonlisation TAgging interface INformation in Services (PERTAINS)
Caroline Williams, Executive Director of Intute [email protected] Andy Priest, Intute Technical Co-ordinator [email protected].
Jan 04, 2016
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Caroline Williams, Executive Director of [email protected]
Andy Priest, Intute Technical [email protected]
PERsonlisation TAgging interface INformation in Services (PERTAINS)
Lisa Charnock
Does your presentation have a title? I'd use it here rather than the conference title if it does.Also - Arial in blue isn't working for me. I'd go with Verdana or Trebuchet, and use colours from the Intute website - I've used a few examples on the slides :). But then....I do get overly preoccupied with fonts!Also - be consistent with your layouts. Do you have a master set up?
Background
•University of Glamorgan and Mimas•Intute and Copac•Follow on from EnTag
PERTAINS – Tag Suggestion Services
• Rationale– Draws on previous work (EnTag project)– Improving quality of user tag metadata
• Basis for suggestions– Using existing document metadata
• Titles and abstracts• Classification terms and existing keywords (user
tags)
– Suggestions from controlled vocabulary• Matching against DDC captions and relative
index terms
Document Metadata• Title• Classification• Uncontrolled keywords • Controlled keywords• Description
Suggestion Services
DDC (SKOS)
• Aim – improve quality of tag metadata• Basis – existing document metadata / DDC (SKOS)• Algorithm – matching, weighting, filtering, ranking• Implementation – URL based web service interface
PERTAINS – Tag Suggestion Services
PERTAINS – Tag Suggestion Services
• Algorithm– Disambiguation of suggestions
• Area of Interest (AOI) filter – restrict suggestions to specific subject areas (DDC summaries)
– Weighting of suggestions - ranking by relevance
• Implementation– Boolean full text matching (MySQL)– URL based web service call interface
• Fast, scalable, platform neutral
– JSON data structures returned
What does it do?
Lisa Charnock
Really like this slide - nice image. I would be inclined to have the image as the hyperlink so you click on the 'answer' and the demo appears.
Related Documents
