Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Browse files

Readme, gitignore, added jar

  • Loading branch information...
commit 6f68718780bad6f5152bff5b0b6a2b8ec0b38f47 1 parent ba6cbbc
@gangeli authored
Showing with 45 additions and 12 deletions.
  1. +1 −0  .gitignore
  2. +34 −0 README
  3. BIN  dist/time.jar
  4. +10 −12 run
View
1  .gitignore
@@ -0,0 +1 @@
+/bin/
View
34 README
@@ -0,0 +1,34 @@
+Parsing Time: Learning to Interpret Time Expressions
+
+----------
+Abstract:
+----------
+We present a probabilistic approach for learning to interpret temporal
+phrases given only a corpus of utterances and the times they reference.
+While most approaches to the task have used regular expressions and similar
+linear pattern interpretation rules, the possibility of phrasal embedding
+and modification in time expressions motivates our use of a compositional
+grammar of time expressions. This grammar is used to construct a latent
+parse which evaluates to the time the phrase would represent, as a logical
+parse might evaluate to a concrete entity. In this way, we can employ a
+loosely supervised EM-style bootstrapping approach to learn these latent
+parses while capturing both syntactic uncertainty and pragmatic ambiguity
+in a probabilistic framework. We achieve an accuracy of 72% on an adapted
+TempEval-2 task -- comparable to state of the art systems.
+
+----------
+Compilation
+----------
+Scala 2.9.1
+JavaNLP April 9 2012
+
+Most experiments can be run with something in the rc file
+ interpret: run the interpretation model (last valid run)
+ interpretAux: run the interpretation model (good parameters)
+ sutimeInterpret: run SUTime
+ heideltimeInterpret: run HeidelTime
+ gutimeInterpret: run GUTime
+
+ detect: run the detection model (last valid run)
+
+run configuration for publication run is in ./run
View
BIN  dist/time.jar
Binary file not shown
View
22 run
@@ -18,15 +18,15 @@ env!(
o('Xrunhprof:cpu=samples,depth=16'),
# o('Xrunhprof:cpu=samples,heap=sites'),
o('server'),
- o('mx10g'),
+ o('mx20g'),
# o('ea'),
'time.Entry',
run(
# -- RUN MODE --
-# o('runInterpretModel'),
+ o('runInterpretModel'),
o('interpretModel', 'aux/interpret.ser.gz'),
- selo(0, 'mode',
+ selo(1, 'mode',
'detect',
'interpret',
'system',
@@ -46,7 +46,7 @@ env!(
'2', #G n^3 + V k logk combined mult/merge (special case of 3)
'3'), #G n^3 + |D_max| k logk lazy evaluation
o('rulePrior', 0.001),
- o('lexPrior', 0.0001),
+ o('lexPrior', 0.00001),
selo(0,'nilPrior',
'lex',
'uniform',
@@ -78,16 +78,15 @@ env!(
o('lexNils'),
o('uniformFirstIteration'),
# <training>
- o('beam', 1500), #1500 works well
+ o('beam', 5000), #1500 works well
o('iters', 20),
o('scoreBeam', 3),
o('maxSearchTime', 10000),
#<detection>
- o('crfFeatureFactory','time.CRFFeatures'),
- o('crfSigma','2.0'),
+ o('crfFeatureFactory','time.TRIPSFeatures'),
#<params>
o('useTime'),
- o('pruneTime', 1000),
+ o('pruneTime', 10000),
o('pruneMinIndex',100),
o('includeTimeProb'),
# o('functionalUnboundedRange'), # BROKEN
@@ -105,7 +104,6 @@ env!(
# -- DATA --
# o('train', 'NYT[0,100]'), o('dev', 'NYT[1000,1100]'),
o('train', 'English[0,162]'), o('dev', 'English[142,162]'),
-# o('train', 'English[20,162]'), o('dev', 'English[0,20]'),
# o('train', 'English[0,142]'), o('dev', 'English[142,162]'),
# o('train', 'English[0,10]'), o('dev', 'English[10,20]'),
# o('train', 'TOY'), o('dev', 'TOY'),
@@ -142,12 +140,12 @@ env!(
# -- BOOTSTRAPING --
#(exec options)
-# o('logDebug'),
+ o('logDebug'),
o('ignoreClasspath', "\"[#{ENV['JAVANLP_HOME']}/.*,etc/scala-.*.jar,etc/postgres.jar,etc/jdom.jar,etc/joda-time.jar]\""),
#(logging)
o('execDir', 'out'),
- o('execOutput', "\"psql://research@localhost:#{JAR}<what?why42?\""),
+ o('execOutput', "\"psql://research@goobs.org:#{JAR}<what?why42?\""),
#(data)
- o('execData', '"psql://research@localhost:data<what?why42?"'),
+ o('execData', '"psql://research@goobs.org:data<what?why42?"'),
nil),
nil)
Please sign in to comment.
Something went wrong with that request. Please try again.