Wednesday, July 1, 2015

Libraries in the semantic web Professor Marcia Lei Zeng

Professor Marcia Lei Zeng Professor, School of Library and Information Science, Kent State University (http://marciazeng.slis.kent.edu) gave a fascinating presentation on Libraries in the semantic web at ANU Library this week.. She challenged us with questions include:


    
  • How can we do more with what we have? 
  • How can we do more with less? [We do not want to buy another system.] 
  • How can we use these LOD data?


In looking at projects she had led focused on turning metadata into linked data she took us through a project on AGRIS that uses controlled vocabulary to link with resources – a brilliant combination of data and text. She suggested that the library catalogue was no longer the central access point and described a flipped model which allowed exploration of external data sources such as DBPedia-Wikipedia, statistical collections and more. She described how hidden access points in the catalogue can bring in much richer information and knowledge through library data, with the example of 5xx fields.

Positing “big text” as the library “bid data” source, she looked at oral history integration through www.unmulitmedia.org which could include hyperlinking automatically in the transcript – an example she used to describe the difference between findability and accessibility. In exploring the concept of the “Internet of things” she used the COGNITO intelligence API as an example of contextual linking of data and opening up of finding aids to connect rich data. She described the work of the Linked Open Data-- Libraries, Archives, and Museums (LAM) Research Group http://lodlam.slis.kent.edu/.

Recent research projects included:
  • Connecting LAMs to the unfamiliar data and metadata resources in the Linked Open Data (LOD) Universe
  •   fact mining, where the team has use automatic semantic analysis tools, testing and compared 3 tools using of 45 archival finding aids drawn from 16 repositories and build the workflow into a tool
  •  Knowledge Organization Systems (KOS) with deliverables including creating local KOS by creating micro-thesaurus, handling special cultural heritage situations (e.g., unknown artists, unidentified creators, un-named objects) for LOD data and Guidelines for creating local name authorities for archival finding aids
  •  Smart big data – with innovation and LCI (Liquid Crystal Institute) researcher-related datasets


She illustrated her big text data use scenarios with the brilliant Nature Video. (2014, July 31). Charting culture https://www.youtube.com/watch?v=4gIhRkCcD4U – do have a look at it.

In summarising her talk she said that we need to embrace the new and changing concepts of the Semantic Web in LAMs – to move:
  •         from "Web of Documents" to "Web of Data”
  •         from linking strings to linking things
  •          from "On the Web" to "Of the Web”
  •          from machine-readable to machine understandable/processable



Professor Marcia Zeng, Ying-Hsang Liu, Charles Sturt University and me

Roxanne Missingham
Australian National University  Librarian


No comments:

Post a Comment