Data Driven Humanities

builderanthologyAI and Robotics

Oct 19, 2013 (3 years and 7 months ago)

82 views

Data Driven Humanities

Gregory Crane

Tufts University

Perseus Digital Library

Two Intertwined Modes


Extending traditional scholarship


Style of Thucydides


Spread of ideas across time/space/culture


Data scholarship

automating the simple


To what Person/Place does
Victoria

refer?


What does
arma virumque cano
mean?


What function does
arma

perform?

Priorities


Automating the simple


necessary

for traditional scholarship


more important
for the humanities


Two grand challenges


Produce more knowledge


Increase the value of existing knowledge

Latin Treebank

Latin Treebank(s)*


Most important advance for the study of
Latin since movable print


Transformative resource for opening up
Latin to non
-
specialists


Syntactic analysis a field unto itself


* NB: US/Italian Collaboration for now…

Three Classes of Technology


From analog to text


ASR, OCR, Machine Vision


From one language to another


MT, CLIR, Translation Support


From text to data


Named entity identification


Information extraction


Text and data mining

Four Classes of Services


Cataloguing services

--

what are the define objects
in our collection?



Named entity services

--

what people, places,
subjects, word senses etc. are
in

those objects?



Customization/Personalization services

--

what do
particular users need at particular times?



Scalable structured contribution services

--

this is
Springfield, IL, not Springfield, MO

1 Cataloguing Services

2 Named Entity Services

3 Personalization Services

Contribution Services

Classicists do get rewarded


A community of tenured faculty


Chris Blackwell,
Furman


Gregory Crane
, Tufts


Helma Dik,
Chicago


Bruce Robertson,
Mount Allison


Jeff Rydberg
-
Cox,
Missouri


Charlotte Rouech
é,
King’s College


Ross Scaife,
Kentucky


Mark S
.,
---


Neel Smith,
Holy Cross

Impact


We need repositories that enable and
support
new

services that
advance

(and don’t just
preserve

existing)
intellectual life.