perceptron features 0.0024276
perceptron model 0.00242648
training algorithm 0.00228931
parsing model 0.002266842
perceptron algorithm 0.00219455
other features 0.002163613
perceptron feature 0.00203358
perceptron training 0.00203054
ing model 0.002012032
linear model 0.001980505
baseline model 0.0019473200000000002
model performance 0.001942483
new features 0.0018940939999999998
language model 0.001882342
generative model 0.0018717690000000001
learning algorithm 0.001833047
base features 0.001806898
model score 0.001803444
useful features 0.001798661
perceptron approach 0.001796899
arbitrary features 0.001793493
global features 0.001781991
ing algorithm 0.001780102
initial model 0.001766864
model scores 0.001735734
training data 0.001725535
ceptron model 0.001721422
punctuation features 0.001711114
tron model 0.001706469
erative model 0.001702549
hybrid model 0.001696174
external model 0.001696174
feature vector 0.0016847329999999999
training sequence 0.001683475
feature set 0.001673955
other perceptron 0.001671793
training set 0.001670915
perceptron parser 0.001669725
algorithm this 0.001657746
parsing approach 0.001637261
training method 0.001603996
training examples 0.001575819
estimation algorithm 0.0015750540000000002
baseline feature 0.00155442
training example 0.001549489
tree pos 0.001513637
same feature 0.001494317
ceptron algorithm 0.001489492
parameter vector 0.00148393
similar feature 0.001471095
refined algorithm 0.001466019
algorithm returns 0.001464461
algorithm intro 0.001464461
training corpus 0.001460794
features 0.00145971
model 0.00145859
baseline perceptron 0.00145662
perceptron models 0.00145255
additional feature 0.001434749
parameter method 0.0014062329999999998
current parameter 0.001387755
first word 0.001386604
ing approach 0.001382451
feature weights 0.0013787439999999999
each training 0.001330811
feature sets 0.001327263
perceptron fea 0.00132198
parse tree 0.001315952
training algo 0.001312989
boolean feature 0.001307772
training sample 0.001307271
training scenario 0.001306172
rable feature 0.001303509
parsing models 0.001292912
parsing performance 0.001292145
parsing accuracy 0.001280142
next word 0.001267747
pos tags 0.001257785
pos tag 0.00125456
parsing this 0.001239338
other words 0.001238135
algorithm 0.00122666
perceptron algo 0.001218229
parameter estimation 0.001213281
pos tagging 0.001201891
test examples 0.001194404
parsing algorithms 0.001187609
parameter updates 0.001186598
pos label 0.001169067
same tree 0.001163969
pos tagger 0.001162981
ing set 0.001161707
parameter values 0.001158381
parsing process 0.001153018
alternative approach 0.001126898
parameter settings 0.0011218
statistical parsing 0.001117266
generative parser 0.0011150140000000001
parameter estima 0.0011117829999999999
incremental parsing 0.00111149
