code features 0.002776947
document features 0.002665209
context features 0.002515674
many features 0.002386728
training document 0.002382819
ing features 0.002370635
training data 0.00236536
training set 0.0022432619999999998
transition features 0.002231415
training test 0.0021872619999999997
emission features 0.002120609
indicator features 0.002059496
generic features 0.002052919
prominent features 0.0020368129999999997
feature set 0.001997162
feature context 0.001987184
training documents 0.00195993
few training 0.001956299
small training 0.001897137
training time 0.001889613
training example 0.001853159
cmc training 0.001828964
training examples 0.001828656
features 0.00182509
much training 0.0018131129999999999
document code 0.0017919759999999998
training points 0.0017759969999999999
enough training 0.001770167
plete training 0.001765619
training instances 0.001758822
ping training 0.001753262
training signal 0.001753262
training mechanism 0.001753262
code set 0.001652419
emission feature 0.001592119
feature vectors 0.0015842809999999999
document text 0.0015661479999999998
training 0.0015427
feature engineering 0.0015339379999999999
code sequence 0.001529029
feature categories 0.00151249
feature templates 0.001508417
test document 0.0014846809999999998
test data 0.001467222
sequence model 0.001399943
sequence data 0.0013998320000000002
document coding 0.00138963
same code 0.001369941
context algorithm 0.001369157
code label 0.001366663
candidate code 0.001358238
test set 0.001345124
document classification 0.001339943
new document 0.001323403
classification model 0.001322595
small code 0.001306294
specific code 0.00129853
feature 0.0012966
possible code 0.001287936
large set 0.001284155
first code 0.0012808659999999999
code terms 0.001278441
final code 0.001274854
previous code 0.0012699299999999998
other codes 0.0012653170000000002
strong code 0.001259879
same document 0.0012582029999999998
code distribution 0.001250551
statistical document 0.001248885
clinical document 0.001245021
code dependency 0.001228602
text classification 0.0012258529999999998
document types 0.001213791
code assignment 0.00121026
code subset 0.00120853
gold code 0.001182581
particular code 0.0011816119999999999
code com 0.00117608
code dependencies 0.0011744239999999999
disease code 0.001173645
code subsets 0.0011704839999999998
distinct code 0.001169037
code semantics 0.001168174
code interactions 0.001166537
code candi 0.001164657
code assignments 0.0011637099999999999
dict code 0.0011623599999999999
input document 0.001160177
example document 0.001150578
coding task 0.001146853
several codes 0.0011411210000000001
clinical text 0.001130931
single model 0.001126712
negative context 0.001115051
triggers document 0.001111674
cmc data 0.001108924
multiclass document 0.00109721
good performance 0.0010924
markov model 0.001091423
local context 0.001086129
