Friday, December 02, 2011

Finding Meaning in Electronic Documents

Greetings from the Sixteenth Australasian Document Computing Symposium (ADCS 2011) at the Australian National University in Canberra. This is being held in conjunction with the 2011 Australasian Language Technology Workshop (ALTA). ADCS is about how to construct, store, manage, use and find electronic documents. But it is also about how to extract meaning from e-documents. That may sound very academic, but this is worth billions of dollars for social media companies (the event is sponsored by Google and NICTA). ADCS is free and there are some spare seats, so if you happen to be in Canberra, and have an interest in the field, come along.

ps: A question I have is why are academic papers eight pages long? Is this due to the mechanics of paper based printing and so no longer relevant, or is eight pages a good size for people to read?

Conference Schedule

9:10Session 1: ADCS/ALTA Joint Session (Chair: David Martinez)

  1. Li Wang, Diana Mccarthy and Timothy Baldwin. Predicting Thread Linking Structure by Lexical Chaining
  2. Diego Molla and Maria Elena Santiago-Martinez. Development of a Corpus for Evidence Based Medicine Summarisation
  3. Mike Symonds, Peter Bruza, Laurianne Sitbon and Ian Turner. Tensor Query Expansion: A cognitively motivated relevance model
  4. Yan Shen, Yuefeng Li, Yue Xu, Renato Lannella, Abdulmohsen Algarni and Xiaohui Tao. An Ontology-based Mining Approach for User Search Intent Discovery
10:40Morning Tea
11:10Session 2 (Chair: Sally Jo Cunningham)

  1. David Alexander, Andrew Trotman and Alistair Knott. Click Log Based Evaluation of Link Discovery
  2. Xiang-Fei Jia, Andrew Trotman and Jason Holdsworth. Fast Search Engine Vocabulary Lookup
  3. (*) Amir Hadad, Alexander Krumpholz, Nina Studeny, Tom Gedeon and David Hawking. Automatic identification of the most important elements in an XML collection
12:00Industry Keynote: Simon Kravis, Fujitsu Australia (Chair: Paul Thomas)

  • Information Retrieval in Large Organisations.
    Information retrieval within large organisations is usually from repositories with little or no hyperlinking of documents. Tools which are successful for public-facing web sites are frequently unsuccessful inside organisations through poor ranking of search results, high levels of duplication and near-duplication and the frequent occurrence of documents which are not completely described by the text which they contain. Metadata-based retrieval is limited by poor coverage from manually assigned metadata and the limited availability of automated metadata assignment.
14:00Session 3 (Chair: Falk Scholer)

  1. Shane Culpepper, Michiko Yasukawa and Falk Scholer. Language Independent Ranked Retrieval with NeWT
  2. David Bainbridge and Sally Jo Cunningham. A Workflow for Document Level Interoperability
  3. Iman Amini, Alexandra Uidtenbogerd and Mark Sanderson. The Interplay of Information Retrieval and Query by Singing With Words
  4. (*) Timothy Jones, Paul Thomas, David Hawking and Ramesh Sankaranarayana. The usefulness of web spam
  5. (*) Su Nam Kim, Timothy Baldwin and Lawrence Cavedon. Harvesting Domain-Specific Terms using Wikipedia
15:20Afternoon Tea
15:50Session 4 (Chair: Andrew Turpin)

  1. Nicholas Sherlock and Andrew Trotman. Efficient sorting of search results by string attributes
  2. Laurence Park. Examining document model residuals to provide feedback during Information Retrieval evaluation
  3. Guido Zuccon, Anthony Nguyen, Teerapong Leelanupab and Leif Azzopardi. Indexing without Spam
  4. (*) Darren Wei-Che Huang, Shlomo Geva, Andrew Trotman and Kelly Y. Itakura. Mobile Applications of Focused Link Discovery
  5. (*) Kelly Itakura and Shlomo Geva. Six Degrees of Cohesion in Patent Citation Network
17:20ADCS Business Meeting

No comments: