The rows are the txt words. The columns are hyp words.
Resource summary:Txt: Horses like apples.
Hyp: Horses like apples. (yes)
Horses NNS |
like VBP |
apples NNS |
|
Horses:NNS | 0.00 | 15.00 | 6.67 |
like:VBP | 15.00 | 0.00 | 13.23 |
apples:NNS | 6.67 | 13.23 | 0.00 |
NO_WORD | 10.00 | 10.00 | 10.00 |
Response: yes (CORRECT)
Justification:
Features matched (wt val name just):
1.00 0.00 Alignment.score
1.00 0.77 Alignment.isGood
-1.00 0.01 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 3.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
Hand-tuned score (dot product of above): 1.1575
Threshold: -2.0000
Txt: The boy saw a bird.
Hyp: The boy saw a comet. (don't know)
The DT |
boy NN |
saw VBD |
a DT |
comet NN |
|
The:DT | 0.00 | 20.00 | 20.00 | 10.00 | 20.00 |
boy:NN | 20.00 | 0.00 | 11.11 | 20.00 | 8.84 |
saw:VBD | 20.00 | 11.11 | 0.00 | 20.00 | 14.99 |
a:DT | 10.00 | 20.00 | 20.00 | 0.00 | 20.00 |
bird:NN | 20.00 | 7.04 | 14.24 | 20.00 | 4.89 |
NO_WORD | 1.00 | 10.00 | 10.00 | 1.00 | 10.00 |
Response: yes (INCORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.98 Alignment.score
1.00 0.56 Alignment.isGood
-1.00 0.03 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 5.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
1.00 1.00 Quant.contract : [a,a]
Hand-tuned score (dot product of above): 1.1485
Threshold: -2.0000
Txt: George bought guns.
Hyp: George bought weapons. (yes)
George NNP |
bought VBD |
weapons NNS |
|
George:NNP | 0.00 | 13.83 | 8.47 |
bought:VBD | 13.83 | 0.00 | 15.00 |
guns:NNS | 9.05 | 15.00 | 1.25 |
NO_WORD | 10.00 | 10.00 | 10.00 |
Response: yes (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.42 Alignment.score
1.00 0.69 Alignment.isGood
-1.00 0.02 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 3.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
1.00 1.00 Hypernym.posWiden : widening in positive context: gun -> weapon
Hand-tuned score (dot product of above): 1.6532
Threshold: -2.0000
Txt: Mars needs women.
Hyp: Mars doesn't need women. (don't know)
Mars NNP |
does VBZ |
n't RB |
need VB |
women NNS |
|
Mars:NNP | 0.00 | 13.00 | 15.50 | 15.50 | 9.51 |
needs:VBZ | 14.39 | 4.29 | 20.00 | 0.00 | 13.09 |
women:NNS | 9.51 | 11.67 | 14.92 | 13.17 | 0.00 |
NO_WORD | 10.00 | 1.00 | 9.00 | 10.00 | 10.00 |
Response: dontknow (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -2.00 Alignment.score
1.00 0.31 Alignment.isGood
-1.00 0.08 Alignment.isBad
-0.10 2.00 Alignment.nbWordNotAligned
0.10 2.00 Alignment.hypSpan
0.10 0.60 Alignment.txtSpan
-6.00 1.00 Adjunct.diffPol : hyp and txt have different polarity
0.00 1.00 NegPolarity.hypNegWord : "need": has child with relation "neg"
0.00 1.00 NegPolarity.hypNegRoot : "need": has child with relation "neg"
-1.00 1.00 NullPunisher.other : n't
-0.05 1.00 NullPunisher.aux : does
Hand-tuned score (dot product of above): -8.7558
Threshold: -2.0000
Txt: Honda manufactures cars.
Hyp: Honda makes cars. (yes)
Honda NNP |
makes VBZ |
cars NNS |
|
Honda:NNP | 0.00 | 15.50 | 10.50 |
manufactures:VBZ | 15.50 | 1.25 | 12.36 |
cars:NNS | 10.50 | 11.67 | 0.00 |
NO_WORD | 10.00 | 10.00 | 10.00 |
Response: yes (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.42 Alignment.score
1.00 0.69 Alignment.isGood
-1.00 0.02 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 3.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
Hand-tuned score (dot product of above): 0.6532
Threshold: -2.0000
Txt: Bush was an excellent student.
Hyp: Bush was a terrible student. (don't know)
Bush NNP |
was VBD |
a DT |
terrible JJ |
student NN |
|
Bush:NNP | 0.00 | 14.07 | 20.50 | 12.50 | 7.72 |
was:VBD | 14.07 | 0.00 | 20.00 | 12.00 | 15.00 |
an:DT | 20.50 | 18.00 | 0.50 | 20.00 | 20.00 |
excellent:JJ | 12.50 | 12.00 | 20.00 | 8.24 | 9.50 |
student:NN | 7.72 | 15.00 | 20.00 | 12.00 | 0.00 |
NO_WORD | 10.00 | 1.00 | 1.00 | 9.00 | 10.00 |
Response: yes (INCORRECT)
Justification:
Features matched (wt val name just):
1.00 -1.75 Alignment.score
1.00 0.37 Alignment.isGood
-1.00 0.06 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 5.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
1.00 1.00 Quant.contract : [a,a]
Hand-tuned score (dot product of above): 0.1596
Threshold: -2.0000
Txt: The United States belongs to the United Nations.
Hyp: The U.S. belongs to the UN. (yes)
The DT |
U.S. NNP |
belongs VBZ |
the DT |
UN NNP |
|
The:DT | 0.00 | 20.50 | 20.00 | 0.00 | 20.50 |
United_States:NNPS | 20.50 | 0.00 | 15.50 | 20.50 | 9.25 |
belongs:VBZ | 20.00 | 15.50 | 0.00 | 20.00 | 15.50 |
the:DT | 0.00 | 20.50 | 20.00 | 0.00 | 20.50 |
United_Nations:NNPS | 20.50 | 7.78 | 15.50 | 20.50 | 0.00 |
NO_WORD | 1.00 | 10.00 | 10.00 | 1.00 | 10.00 |
Response: yes (CORRECT)
Justification:
Features matched (wt val name just):
1.00 0.00 Alignment.score
1.00 0.77 Alignment.isGood
-1.00 0.01 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 5.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
Hand-tuned score (dot product of above): 1.3575
Threshold: -2.0000
Txt: Two senators objected.
Hyp: 14 senators objected. (don't know)
14 CD |
senators NNS |
objected VBD |
|
Two:CD | 5.00 | 20.50 | 20.50 |
senators:NNS | 20.42 | 0.00 | 12.58 |
objected:VBD | 20.32 | 12.58 | 0.00 |
NO_WORD | 10.00 | 10.00 | 10.00 |
Response: dontknow (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -1.67 Alignment.score
1.00 0.39 Alignment.isGood
-1.00 0.06 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 3.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
-6.00 1.00 Numeric.mismatch : NUMBER mismatch: '14.0' vs '2.0'
Hand-tuned score (dot product of above): -6.9368
Threshold: -2.0000
Txt: The victim was a homeless immigrant.
Hyp: The victim was an immigrant. (yes)
The DT |
victim NN |
was VBD |
an DT |
immigrant JJ |
|
The:DT | 0.00 | 20.00 | 20.00 | 10.00 | 20.00 |
victim:NN | 20.00 | 0.00 | 15.00 | 20.00 | 10.40 |
was:VBD | 20.00 | 15.00 | 0.00 | 18.00 | 12.00 |
a:DT | 10.00 | 20.00 | 20.00 | 0.50 | 20.00 |
homeless:JJ | 20.00 | 8.96 | 12.00 | 20.00 | 6.73 |
immigrant:NN | 20.00 | 8.28 | 15.00 | 20.00 | 0.00 |
NO_WORD | 1.00 | 10.00 | 1.00 | 1.00 | 9.00 |
Response: yes (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.10 Alignment.score
1.00 0.75 Alignment.isGood
-1.00 0.01 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 5.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
0.50 1.00 Adjunct.dropPosCxt : text adjunct "homeless" of "immigrant" dropped on aligned hyp word "immigrant"
1.00 1.00 Quant.contract : [a,a]
0.50 1.00 Adjunct.dropPosCxt&Align.veryGood :
Hand-tuned score (dot product of above): 3.2381
Threshold: -2.0000
Txt: The deal included power plants.
Hyp: The deal included nuclear power plants. (don't know)
The DT |
deal NN |
included VBD |
nuclear_power NNS |
plants NNS |
|
The:DT | 0.00 | 20.00 | 20.00 | 20.00 | 20.00 |
deal:NN | 20.00 | 0.00 | 15.00 | 9.76 | 7.94 |
included:VBD | 20.00 | 15.00 | 0.00 | 14.52 | 15.00 |
power_plants:NNS | 20.00 | 9.17 | 15.00 | 6.30 | 0.00 |
NO_WORD | 1.00 | 10.00 | 10.00 | 10.00 | 10.00 |
Response: dontknow (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -2.00 Alignment.score
1.00 0.31 Alignment.isGood
-1.00 0.08 Alignment.isBad
-0.10 1.00 Alignment.nbWordNotAligned
0.10 3.00 Alignment.hypSpan
0.10 0.80 Alignment.txtSpan
-1.00 1.00 Adjunct.addPosCxt : the hypothesis added the modifier word "nuclear_power"
-1.00 1.00 NullPunisher.other : nuclear_power
Hand-tuned score (dot product of above): -3.4858
Threshold: -2.0000
Txt: The Taj Mahal is in India.
Hyp: The Taj Mahal is in Asia. (yes)
The DT |
Taj_Mahal NNP |
is VBZ |
Asia NNP |
|
The:DT | 0.00 | 20.50 | 20.00 | 20.50 |
Taj_Mahal:NNP | 20.50 | 0.00 | 15.50 | 10.00 |
is:VBZ | 20.00 | 15.50 | 0.00 | 15.50 |
India:NNP | 20.50 | 10.00 | 15.50 | 6.67 |
NO_WORD | 1.00 | 10.00 | 10.00 | 10.00 |
Response: yes (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -1.67 Alignment.score
1.00 0.39 Alignment.isGood
-1.00 0.06 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 4.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
Hand-tuned score (dot product of above): -0.8368
Threshold: -2.0000
Txt: Monet was born in Paris.
Hyp: Monet was born in Italy. (don't know)
Monet NNP |
was VBD |
born VBN |
Italy NNP |
|
Monet:NNP | 0.00 | 15.50 | 14.39 | 9.76 |
was:VBD | 15.50 | 0.00 | 6.03 | 15.50 |
born:VBN | 14.39 | 6.03 | 0.00 | 15.50 |
Paris:NNP | 9.75 | 13.00 | 14.39 | 7.02 |
NO_WORD | 10.00 | 1.00 | 10.00 | 10.00 |
Response: yes (INCORRECT)
Justification:
Features matched (wt val name just):
1.00 -1.76 Alignment.score
1.00 0.36 Alignment.isGood
-1.00 0.06 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 4.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
Hand-tuned score (dot product of above): -0.9516
Threshold: -2.0000
Txt: Talks will resume on August 14.
Hyp: Talks will resume in August. (yes)
Talks NNS |
will MD |
resume VB |
August NNP |
|
Talks:NNS | 0.00 | 18.89 | 15.00 | 8.81 |
will:MD | 18.89 | 0.00 | 18.01 | 20.50 |
resume:VB | 15.00 | 18.01 | 0.00 | 13.83 |
August:NNP | 8.81 | 20.50 | 13.83 | 0.00 |
14:CD | 20.50 | 20.50 | 20.32 | 20.00 |
NO_WORD | 10.00 | 10.00 | 10.00 | 10.00 |
Response: yes (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.25 Alignment.score
1.00 0.72 Alignment.isGood
-1.00 0.01 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 4.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
0.50 1.00 Adjunct.dropPosCxt : text adjunct "14" of "August" dropped on aligned hyp word "August"
1.00 1.00 Date.matchDatesByGraph : hyp/txt matching, by graph: August and children
-1.00 1.00 Structure.relMismatch : text "August" is prep_on of "resume" while hyp "August" is prep_in of "resume" which aligned to text "resume"
0.50 1.00 Adjunct.dropPosCxt&Align.veryGood :
Hand-tuned score (dot product of above): 1.9571
Threshold: -2.0000
Txt: Hiroshima was bombed on August 6.
Hyp: Hiroshima was bombed on August 8. (don't know)
Hiroshima NNP |
was VBD |
bombed VBN |
August NNP |
8 CD |
|
Hiroshima:NNP | 0.00 | 15.50 | 15.50 | 9.85 | 20.50 |
was:VBD | 15.50 | 0.00 | 7.41 | 15.50 | 20.50 |
bombed:VBN | 15.50 | 7.41 | 0.00 | 15.50 | 20.04 |
August_6:NNP | 9.85 | 15.50 | 15.23 | 0.00 | 11.30 |
NO_WORD | 10.00 | 1.00 | 10.00 | 10.00 | 10.00 |
Response: dontknow (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -2.00 Alignment.score
1.00 0.31 Alignment.isGood
-1.00 0.08 Alignment.isBad
-0.10 1.00 Alignment.nbWordNotAligned
0.10 4.00 Alignment.hypSpan
0.10 0.80 Alignment.txtSpan
-1.00 1.00 Adjunct.addPosCxt : the hypothesis added the modifier word "8"
-3.00 1.00 Date.dateHeadMismatch : August vs. August_6
-3.00 1.00 NullPunisher.entity : 8
Hand-tuned score (dot product of above): -8.3858
Threshold: -2.0000
Txt: North Korea succeeded in testing a nuclear device.
Hyp: North Korea tested a nuclear device. (yes)
North_Korea NNP |
tested VBD |
a DT |
nuclear JJ |
device NN |
|
North_Korea:NNP | 0.00 | 15.50 | 20.50 | 12.50 | 10.50 |
succeeded:VBD | 15.50 | 6.79 | 20.00 | 9.50 | 15.00 |
in:IN | 20.50 | 20.00 | 13.83 | 20.00 | 20.00 |
testing:VBG | 15.50 | 0.00 | 20.00 | 8.23 | 8.26 |
a:DT | 20.50 | 20.00 | 0.00 | 20.00 | 20.00 |
nuclear:JJ | 12.50 | 8.96 | 20.00 | 0.00 | 9.68 |
device:NN | 10.50 | 9.30 | 20.00 | 9.68 | 0.00 |
NO_WORD | 10.00 | 10.00 | 1.00 | 9.00 | 10.00 |
Response: dontknow (INCORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.40 Alignment.score
1.00 0.69 Alignment.isGood
-1.00 0.02 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 5.00 Alignment.hypSpan
0.10 0.80 Alignment.txtSpan
1.00 1.00 Quant.contract : [a,a]
-3.00 1.00 Structure.argsMismatch : args have different parents but same relations: text "North_Korea" <-nsubj-- "succeeded" vs. hyp "North_Korea" <-nsubj-- "tested", which aligned to text "testing"
-3.00 1.00 Structure.argsMismatch&Align.veryGood :
Hand-tuned score (dot product of above): -4.1463
Threshold: -2.0000
Txt: The administration denied placing illegal wiretaps.
Hyp: The administration placed illegal wiretaps. (don't know)
The DT |
administration NN |
placed VBD |
illegal JJ |
wiretaps NNS |
|
The:DT | 0.00 | 20.00 | 20.00 | 20.00 | 20.00 |
administration:NN | 20.00 | 0.00 | 14.96 | 9.44 | 6.85 |
denied:VBD | 20.00 | 14.35 | 6.67 | 7.80 | 12.56 |
placing:VBG | 20.00 | 14.68 | 0.00 | 9.62 | 13.38 |
illegal:JJ | 20.00 | 9.44 | 11.23 | 0.00 | 7.97 |
wiretaps:NNS | 20.00 | 6.85 | 15.00 | 7.97 | 0.00 |
NO_WORD | 1.00 | 10.00 | 10.00 | 9.00 | 10.00 |
Response: dontknow (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.40 Alignment.score
1.00 0.69 Alignment.isGood
-1.00 0.02 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 5.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
-3.00 1.00 Structure.argsMismatch : args have different parents but same relations: text "administration" <-nsubj-- "denied" vs. hyp "administration" <-nsubj-- "placed", which aligned to text "placing"
-3.00 1.00 Structure.argsMismatch&Align.veryGood :
Hand-tuned score (dot product of above): -5.1263
Threshold: -2.0000
Txt: Antonio Fazio, governor of the Bank of Italy, is engulfed in scandal.
Hyp: Antonio Fazio is governor of the Bank of Italy. (yes)
Antonio_Fazio NNP |
is VBZ |
governor NN |
the DT |
Bank_of_Italy NNP |
|
Antonio_Fazio:NNP | 0.00 | 15.50 | 10.50 | 20.50 | 8.71 |
governor:NN | 0.50 | 14.23 | 0.00 | 19.23 | 8.90 |
the:DT | 20.50 | 20.00 | 20.00 | 0.00 | 20.50 |
Bank_of_Italy:NNP | 8.71 | 15.50 | 8.90 | 20.50 | 0.00 |
is:VBZ | 15.50 | 0.00 | 15.00 | 20.00 | 15.50 |
engulfed:VBN | 15.50 | 6.35 | 14.95 | 20.00 | 15.50 |
scandal:NN | 10.50 | 15.00 | 8.40 | 20.00 | 9.17 |
NO_WORD | 10.00 | 1.00 | 10.00 | 1.00 | 10.00 |
Response: dontknow (INCORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.60 Alignment.score
1.00 0.65 Alignment.isGood
-1.00 0.02 Alignment.isBad
-0.10 1.00 Alignment.nbWordNotAligned
0.10 3.00 Alignment.hypSpan
0.10 0.80 Alignment.txtSpan
-0.05 1.00 NullPunisher.aux : is
-2.00 1.00 Structure.parentsMismatch : args have different parents, different relations: text "Antonio_Fazio" <-nsubjpass-- "engulfed" vs. hyp "Antonio_Fazio" <-nsubj-- "governor", which aligned to text "governor"
-3.00 1.00 Structure.parentsMismatch&Align.veryGood :
Hand-tuned score (dot product of above): -4.7442
Threshold: -2.0000
Txt: Lincoln was assassinated by John Wilkes Booth.
Hyp: Lincoln assassinated John Wilkes Booth. (don't know)
Lincoln NNP |
assassinated VBD |
John_Wilkes_Booth NNP |
|
Lincoln:NNP | 0.00 | 15.50 | 8.40 |
was:VBD | 15.50 | 7.98 | 15.50 |
assassinated:VBN | 15.50 | 0.00 | 15.50 |
John_Wilkes_Booth:NNP | 8.40 | 15.50 | 0.00 |
NO_WORD | 10.00 | 10.00 | 10.00 |
Response: dontknow (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -1.33 Alignment.score
1.00 0.47 Alignment.isGood
-1.00 0.04 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 3.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
-3.00 1.00 Structure.clearBadness : for predicate assassinated, text actor set [John_Wilkes_Booth] overlaps hyp undergoer set [John_Wilkes_Booth]
-3.00 1.00 Structure.clearBadness : for predicate assassinated, text undergoer set [John_Wilkes_Booth] overlaps hyp actor set [John_Wilkes_Booth]
-1.00 1.00 Structure.relMismatch : text "Lincoln" is nsubjpass of "assassinated" while hyp "Lincoln" is nsubj of "assassinated" which aligned to text "assassinated" text "John_Wilkes_Booth" is agent of "assassinated" while hyp "John_Wilkes_Booth" is dobj of "assassinated" which aligned to text "assassinated"
Hand-tuned score (dot product of above): -7.5071
Threshold: -2.0000
Txt: Spain sold warplanes to Venezuela.
Hyp: Venezuela bought warplanes from Spain. (yes)
Venezuela NNP |
bought VBD |
warplanes NNS |
Spain NNP |
|
Spain:NNP | 6.03 | 15.50 | 9.70 | 0.00 |
sold:VBD | 15.50 | 0.00 | 15.00 | 14.39 |
warplanes:NNS | 9.76 | 14.52 | 0.00 | 9.70 |
Venezuela:NNP | 0.00 | 15.50 | 9.76 | 6.03 |
NO_WORD | 10.00 | 10.00 | 10.00 | 10.00 |
Response: yes (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -1.00 Alignment.score
1.00 0.55 Alignment.isGood
-1.00 0.03 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 4.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
Hand-tuned score (dot product of above): 0.0205
Threshold: -2.0000
Txt: Brown criticized Blair sharply.
Hyp: Blair criticized Brown sharply. (don't know)
Blair NNP |
criticized VBD |
Brown NNP |
sharply RB |
|
Brown:NNP | 8.00 | 15.50 | 0.00 | 15.50 |
criticized:VBD | 15.50 | 0.00 | 15.50 | 18.78 |
Blair:NNP | 0.00 | 15.50 | 8.00 | 15.50 |
sharply:RB | 15.50 | 18.78 | 15.50 | 0.00 |
NO_WORD | 10.00 | 10.00 | 10.00 | 9.00 |
Response: dontknow (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -1.00 Alignment.score
1.00 0.55 Alignment.isGood
-1.00 0.03 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 4.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
-3.00 1.00 Structure.clearBadness : for predicate criticized, text actor set [Brown] overlaps hyp undergoer set [Brown]
-3.00 1.00 Structure.clearBadness : for predicate criticized, text undergoer set [Brown] overlaps hyp actor set [Brown]
-1.00 1.00 Structure.relMismatch : text "Blair" is dobj of "criticized" while hyp "Blair" is nsubj of "criticized" which aligned to text "criticized" text "Brown" is nsubj of "criticized" while hyp "Brown" is dobj of "criticized" which aligned to text "criticized"
Hand-tuned score (dot product of above): -6.9795
Threshold: -2.0000
Txt: Spain sells olive oil to Italy.
Hyp: Italy lost to Spain in the World Cup. (don't know)
Italy NNP |
lost VBD |
Spain NNP |
the DT |
World_Cup NNP |
|
Spain:NNP | 3.17 | 15.50 | 0.00 | 20.50 | 10.50 |
sells:VBZ | 13.50 | 7.68 | 13.50 | 20.00 | 15.50 |
olive_oil:NN | 10.50 | 14.70 | 10.50 | 20.00 | 9.39 |
Italy:NNP | 0.00 | 15.50 | 3.17 | 20.50 | 10.50 |
NO_WORD | 10.00 | 10.00 | 10.00 | 1.00 | 10.00 |
Response: dontknow (CORRECT)
Justification:
Features matched (wt val name just):
1.00 -4.54 Alignment.score
1.00 0.03 Alignment.isGood
-1.00 0.51 Alignment.isBad
-0.10 2.00 Alignment.nbWordNotAligned
0.10 3.00 Alignment.hypSpan
0.10 0.40 Alignment.txtSpan
-1.00 1.00 Adjunct.addPosCxt : the hypothesis added the modifier word "World_Cup"
-0.10 1.00 NullPunisher.article : the
-3.00 1.00 NullPunisher.entity : World_Cup
-1.00 1.00 Structure.relMismatch : text "Italy" is prep_to of "sells" while hyp "Italy" is nsubj of "lost" which aligned to text "sells" text "Spain" is nsubj of "sells" while hyp "Spain" is prep_to of "lost" which aligned to text "sells"
Hand-tuned score (dot product of above): -9.9706
Threshold: -2.0000
Txt: The bomb exploded on August 8 as experts had warned.
Hyp: The bomb exploded as experts had warned on August 8. (don't know)
The DT |
bomb NN |
exploded VBD |
as IN |
experts NNS |
had VBD |
warned VBN |
August NNP |
8 CD |
|
The:DT | 0.00 | 20.00 | 20.00 | 20.00 | 20.00 | 20.00 | 20.00 | 20.50 | 20.50 |
bomb:NN | 20.00 | 0.00 | 8.15 | 20.00 | 8.03 | 15.00 | 12.52 | 9.66 | 19.31 |
exploded:VBD | 20.00 | 8.15 | 0.00 | 20.00 | 11.67 | 8.06 | 7.91 | 15.50 | 19.80 |
August:NNP | 20.50 | 9.66 | 15.50 | 20.50 | 8.11 | 15.50 | 15.50 | 0.00 | 20.00 |
8:CD | 20.50 | 19.31 | 19.80 | 20.50 | 20.36 | 20.50 | 20.36 | 20.00 | 0.00 |
as:IN | 20.00 | 20.00 | 20.00 | 0.00 | 20.00 | 18.00 | 20.00 | 20.50 | 20.50 |
experts:NNS | 20.00 | 8.03 | 11.67 | 20.00 | 0.00 | 15.00 | 14.29 | 8.11 | 20.36 |
had:VBD | 20.00 | 15.00 | 8.06 | 18.00 | 15.00 | 0.00 | 6.19 | 15.50 | 20.50 |
warned:VBN | 20.00 | 12.52 | 7.91 | 20.00 | 14.29 | 6.19 | 0.00 | 15.50 | 20.36 |
NO_WORD | 1.00 | 10.00 | 10.00 | 10.00 | 10.00 | 1.00 | 10.00 | 10.00 | 10.00 |
Response: yes (INCORRECT)
Justification:
Features matched (wt val name just):
1.00 -0.22 Alignment.score
1.00 0.73 Alignment.isGood
-1.00 0.01 Alignment.isBad
-0.10 0.00 Alignment.nbWordNotAligned
0.10 9.00 Alignment.hypSpan
0.10 1.00 Alignment.txtSpan
0.50 1.00 Date.matchDatesByNormForm : hyp/txt matching, by normalized form: 08/08/1000
Hand-tuned score (dot product of above): 1.9908
Threshold: -2.0000
Hand-set weights Accuracy: 16/22 = 0.7273
Word similarity table built on Tue Mar 13 13:08:42 PDT 2007 using command:
java edu.stanford.nlp.rte.WordSimilarityGenerator -info /u/nlp/rte/data/byformat/align/simple/simple-test.align.xml -output /u/nlp/rte/data/byformat/wordsim/simple/simple-test.wordsim.html -alignmentWeights manual