other corpus 0.003502339
training corpus 0.0032484569999999997
irex corpus 0.003002587
corpus table 0.002996114
paper corpus 0.002983783
unannotated corpus 0.002969222
enough corpus 0.002958164
annotated corpus 0.0029450329999999997
nhk corpus 0.002925039
notated corpus 0.002923632
corpus 0.0026395
feature character 0.001565863
training data 0.0015066720000000001
chinese character 0.0014222240000000001
ing data 0.0013843
morpheme feature 0.001378541
such kind 0.0013043030000000001
novel method 0.0012901079999999999
robust method 0.001281227
proposed method 0.001269927
machine learning 0.001251296
character type 0.001228259
similar morpheme 0.001213015
unfamiliar word 0.001201892
familiar word 0.001194198
type feature 0.00118597
japanese language 0.0011854209999999999
learning approaches 0.001184223
unfamiliar words 0.001174057
hiragana character 0.001098804
chine learning 0.001080945
english pos 0.00103734
chunk label 0.001036447
pos type 0.001001705
similar morphemes 9.99699E-4
label translation 9.86021E-4
method 9.83503E-4
translation rules 9.57306E-4
japanese nes 9.46755E-4
unfamiliar morpheme 9.30354E-4
conditional random 9.274330000000001E-4
similarity function 9.24514E-4
familiar morpheme 9.2266E-4
training instance 9.16887E-4
ing nes 9.09055E-4
morpheme mˆu 9.05447E-4
ilar morpheme 9.00708E-4
large database 8.961659999999999E-4
first step 8.83373E-4
boundary problem 8.81761E-4
context vector 8.7424E-4
morphological analyzer 8.712749999999999E-4
organization names 8.60981E-4
words 8.60457E-4
second step 8.56812E-4
january japanese 8.53879E-4
chunk labels 8.397579999999999E-4
type frequency 8.396510000000001E-4
vector machine 8.32224E-4
random fields 8.315200000000001E-4
above problem 8.30434E-4
gen particle 8.247999999999999E-4
surface string 8.173189999999999E-4
newspaper articles 8.158029999999999E-4
top particle 8.133299999999999E-4
decision list 8.12337E-4
company names 8.06743E-4
character 8.04076E-4
context vectors 8.013709999999999E-4
possible unigrams 8.00547E-4
learning 7.97004E-4
various nlp 7.96738E-4
personal names 7.927629999999999E-4
segmentation boundary 7.91655E-4
support vector 7.85152E-4
face string 7.84905E-4
original morphemes 7.80899E-4
experimental evaluation 7.70863E-4
traditional machine 7.66217E-4
type chunk 7.65878E-4
feature 7.61787E-4
type fea 7.55909E-4
unannotated cor 7.513929999999999E-4
english alphabet 7.48424E-4
ditional machine 7.40297E-4
tor machine 7.40297E-4
baseline methods 7.39136E-4
problematic morphemes 7.37238E-4
features 7.26464E-4
important sub 7.25285E-4
unfamiliar morphemes 7.170379999999999E-4
few times 7.15985E-4
tract nes 7.112329999999999E-4
familiar morphemes 7.09344E-4
cosine function 7.055970000000001E-4
support vec 7.046680000000001E-4
cross validation 7.04028E-4
tjong kim 7.032E-4
kim sang 7.032E-4
parsing direction 7.0135E-4
