Saturday 14 Dec 2019
  • Increase font size
  • Default font size
  • Decrease font size
  • default style
  • blue style
  • green style
  • red style
  • orange style
You are here:

6th - The Correspondence Analysis Platform for Uncovering Deep Structure in Data and Information: Fionn Murtagh, SFI

6th Annual Public Boole Lecture in Informatics

Title: "The Correspondence Analysis Platform for Uncovering Deep Structure

in Data and Information"

 

Speaker: Professor  Fionn Murtagh

Director

Information, Communications & Emergent Technologies

Science Foundation Ireland

 

Time & Date: 8pm on Tuesday 29th April 2008  (refreshments available from 7.30 pm)

Location: Boole Lecture Theatre 3, UCC

 

THIS YEAR APPLE HAS KINDLY DONATED AN IPOD TOUCH WHICH MUST BE WON BY ONE LUCKY ATTENDEE

Our work is based on two different aspects of information semantics:

  1. the collection of all relationships,
  2. tracking and spotting anomaly and change.

 

As we will exemplify, the first is implemented by endowing all relevant information spaces with a Euclidean metric in a common projected space.  The second is modeled by an induced ultrametric.

A very general way to achieve a Euclidean embedding of different information spaces based on cross-tabulation counts (and from other data) is provided by Correspondence Analysis.  From there, the induced ultrametric that we are particularly interested in takes a sequential - e.g. temporal - ordering of the data into account.

 

We employ such a perspective to look at narrative, "the flow of thought and the flow of language" in W. Chafe's description.  An ideal template for many domains of application is provided by filmscript.  Taking Robert McKee's book on "Story: Substance, Structure, Style and the Principles of Screenwriting" as our framework, we show how many qualitative perspectives can be converted into a quantitative analysis of semantics.

 

In a short review of the global economic conjuncture related to Ireland's Celtic Tiger and more recent period, based on the annual Taulbee Survey Report of the Computing Research Association, we address two preliminary issues that are important for the analysis of deep structure in technology trends: Do we have a mechanism available to focus description and characterization of these trends in a small number of dimensions?  And arising from two significant recent technology upswings, can we suggest the basis for the next one?

ADMISSION IS FREE & EVERYONE is welcome to attend

 

Photos of the Sixth Annual Boole Lecture in Informatics 29th April 2008