Difference between revisions of "Iarpa"
From Knowitall
(→TODO) |
|||
Line 1: | Line 1: | ||
+ | == Links == | ||
+ | |||
+ | * https://jira.esg.sri.com/ | ||
+ | * https://filex.ai.sri.com/dovetail/ | ||
+ | * https://svn.ai.sri.com/projects/dovetail/trunk | ||
+ | * https://wiki.sri.com/ | ||
+ | |||
== TODO == | == TODO == | ||
=== June 15 === | === June 15 === |
Revision as of 18:37, 19 April 2012
Contents
Links
- https://jira.esg.sri.com/
- https://filex.ai.sri.com/dovetail/
- https://svn.ai.sri.com/projects/dovetail/trunk
- https://wiki.sri.com/
TODO
June 15
- Compare Janara's work with nesty
- Remove WEKA from R2A2
- Extend relational noun to plurals
Engineering Tasks
- Handle capitalized/allcaps/nocap sentences.
- Oddity of data where person lastname is in parentheses.
- Handle geotagging and other oddities of the classified data.
- Address efficiency issues.
Research Tasks
- Pronoun Resolution
Tools
Speed
- Compile to native code.
- Compare NLP libraries.
Conversion of Docs to Text
- Detect headlines and handle separately
- Sentences with only single letters
- Detect bullet points