training data 0.001793532
input output 0.0017733480000000001
constituent input 0.0016765349999999998
output constituent 0.001634761
other words 0.001600025
word tags 0.001591421
parse data 0.001589425
output information 0.0014918840000000002
same constituent 0.0014688219999999998
training set 0.001466843
ing data 0.001460574
data set 0.001457477
new input 0.0014105300000000001
learning algorithm 0.0013853490000000001
input tags 0.001377942
training sentences 0.001369641
new output 0.001368756
input units 0.001337148
next input 0.0013264940000000001
syntactic parser 0.001323303
test data 0.001306398
small training 0.001303629
output units 0.001295374
output con 0.0012821759999999999
infrequent words 0.0012777449999999998
input activation 0.001276913
new constituent 0.001271943
input tag 0.001268797
output unit 0.001264931
target output 0.001264907
example input 0.001257929
only input 0.001253034
data problems 0.001238037
first constituent 0.001228457
input sequence 0.001224254
corpus constituent 0.001215434
input pattern 0.001199222
such networks 0.001197798
parse tree 0.001193794
long training 0.001190123
single constituent 0.001188986
gestalt input 0.0011859729999999999
second constituent 0.001184274
network con 0.0011822059999999999
different sentence 0.001182145
sparse data 0.0011769789999999999
constituent context 0.001172901
other constituents 0.001165643
validation data 0.00116315
ing algorithm 0.001161779
fixed input 0.001159388
training sets 0.001158018
output pattern 0.0011574480000000002
training cross 0.001155053
data sets 0.001148652
training times 0.001148651
learning algorithms 0.001139715
mental output 0.0011390950000000001
training methodology 0.00113797
other set 0.0011369190000000001
quent input 0.001135518
same units 0.001129435
expt training 0.001129311
syntactic parsing 0.001126581
fixed output 0.001117614
sentence position 0.001113702
connectionist network 0.001111638
sibling output 0.001111485
output pat 0.001107973
network performance 0.001107201
many constituents 0.001106542
grandparent output 0.001095668
accumulated output 0.001095668
ling output 0.001093701
output mappings 0.001093701
current parse 0.001091016
synchrony network 0.001085403
linguistic constituents 0.001082946
constituent labels 0.00107749
child constituent 0.001067646
learning experiments 0.0010674970000000002
correct parse 0.001065749
several constituents 0.001061458
particular constituent 0.00105244
parse trees 0.001049781
ing structure 0.001048085
example parse 0.0010477099999999999
constituent inputs 0.001046075
learning abilities 0.0010399950000000002
ach constituent 0.0010380139999999999
backpropagation learning 0.001036815
same method 0.001032867
recurrent network 0.001031195
effective learning 0.001031108
words 0.0010285
parent constituent 0.0010215889999999998
parsing systems 0.001019302
tionist network 0.001012834
same head 0.0010127439999999999
constituent memory 0.0010051399999999999
