NAER-101-12-F-2-03-00-3-02

101 1 101 12
101 12
NAER

101

12

F

2

03

00

3

02



................................................................................1

........................................................................................................2

............................................................................................3

(concordancer) .................................................................3

..........................................................................................................4

(lemmatizer)...................................................................5

....................................................................................................5

(part-of-speech tagger)...............................................................5

(parser) ...........................................................................................6

....................................................................................................9

......................................................................................10

..........................................................................................12

(Corpus Processing) .............................................13

tr ..............................................................................................13

sort...................................................................................................13

uniq ........................................................................13

grep .....................................14

awk .........................................................14

perl ...................................................................................................................14

......................................................................................15

(mutual information)........................................................................15

t (t-score) ............................................................................................................ 15

(entropy) ............................................................................................................... 16

n (ngram language model) ............................................................ 16

..................................................................................17

i

 ..........................................................................................21

..............................................................................21

......................................................22

() .......................................................................23

() ...................................................................................................23

() Champollion Tool Kit (CTK) .........................27

..........................................................29

NP Chunking............................................................34

................................................42

........................................................................................46

Lucene ..............................................................49

LDC Sketch Engine .........................50

........................................................................................................50

........................................................................................................................................ 53

............................................................... 53

Support Vector Machine (SVM) ............................. 55

Bayesian Classification ..................................................................... 57

................................................................................................................................ 58

................................................................................................................................ 59

ii


AntConc ...............................................................4

AntConc ...............................................................................4

......................................6

......................................7

..........................................7

Stanford Parser ....................................................8

Stanford Parser ....................................................8

Lund Mate-too ...................9

Lund Mate-tool ..................9

....................................................10

................................................10

E-HowNet .........................................................11

........................................................17

....................................................18

() .................................18

....................................................19

....................19

Sketch Engine ................................................................20

Sketch Engine ........................................................20

CTK ......................28

............................................................28

........................................................................36

Yamcha .....................43

Lucene ...................................................49

iii



1947 Shannon (Noise

Chanel Model)(information theory)

1961

Francis Kucera Brown Corpus

Brown Corpus

Noam Chomsky

1960 1980 20

1980

(Hidden Markov Model)

John SinclairGeoffrey Leech, Sidney Greenbaum,

Jan Svartvik, Randolph Quirk.

Sinclair (1987)1, Quirk, Greenbaum, Leech, Svartvik (1985)2, Garside,

Leech, Sampson (1987)3 John Sinclair Harper Collins

1980 Collins

Cobuild Collins Cobuild

1
Looking Up: An Account of the Cobuild Project.
2
A Comprehensive Grammar of the English Language.
3
The Computational Analysis of English : A Corpus-Based Approach.
1



1990

(automatic

lexical knowledge acquisition)

1990

Brown Corpus, LOB Lancaster-Oslo-Bergen) Corpus, BNC (British National

Corpus), Project Gutenburg

Penn Corpus, Sussane Corpus

( 500 )

Penn Treebank

Hansard Corpus

Sinica Treebank (http://www.aclclp.org.tw/use_stb_c.php)

2

Penn Chinese Treebank

(http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2004T05)

Sinica Treebank Penn Chinese Treebank

PP, NP

FRAG (sentence)( IP )

Sinica Treebank ( Head-Driven Principle )

(Head)

Penn Chinese Treebank

SBJ OBJ

Sinica Treebank

Penn Treebank Sinica Treebank

Sinica Treebank Penn Chinese

Treebank Linguistic Data Consortium (LDC) Penn

Chinese Treebank Chinese Proposition Bank

(concordancer)

Antconc

Antconc concordancer

3

 AntConc
http://www.antlab.sci.waseda.ac.jp/antconc_index.html

Antconc concordancer Concordance hits

Antconc Wordlist

AntConc

4

http://www.antlab.sci.waseda.ac.jp/antconc_index.html

(lemmatizer)

90 97

LingPipe

http://alias-i.com/lingpipe/demos/tutorial/chineseTokens/read-me.html

Chinese Word Segmenter http://nlp.stanford.edu/software/segmenter.shtml

(part-of-speech tagger):

98, Stanford Parser

Chinese Word Segmenter

(tagset)

5


http://ckipsvr.iis.sinica.edu.tw/

(parser)

(partial parse)

, Minipar Stanford Parser Lund

Mate-tool

Sinica Treebank Penn Chinese Treebank

PP, NP

FRAG (sentence)( IP )

Sinica Treebank ( Head-Driven Principle )

(Head)

Penn Chinese Treebank

SBJ OBJ

terminal node

head(node)

Stanford Parser Stanford Parser

NP VP IP

7


Stanford Parser
http://nlp.stanford.edu:8080/parser/

Stanford Parser
http://nlp.stanford.edu:8080/parser/

8

 Lund Mate-tool
http://barbar.cs.lth.se:8091/

Lund Mate-tool
http://barbar.cs.lth.se:8091/

9


http://dict.revised.moe.edu.tw/

http://dict.revised.moe.edu.tw/

Hownet

( Dong and Dong (2006) Hownet

10

 Hownet doctor, surgeon,

doctor {human| :HostOf={Occupation|

},domain={medical|},{doctor|:agent={~}}}

meta language,

{doctor|}

HowNet E-HowNet

E-HowNet
http://ehownet.iis.sinica.edu.tw/

Ae142 Ae151

Ae142","" "Ae142","" "Ae142","" "Ae142",""

"Ae142"," " "Ae142"," " "Ae142"," " "Ae142"," "

"Ae151","" "Ae151","" "Ae151","" "Ae151","" "Ae151",""

"Ae151","" "Ae151","" "Ae151","" "Ae151","" "Ae151","

11

" "Ae151","" "Ae151","" "Ae151","" "Ae151","" "Ae151","

" "Ae151","" "Ae151","" "Ae151","" "Ae151","" "Ae151",""

"Ae151","" "Ae151","" "Ae151","" "Ae151","" "Ae151",""

"Ae151"," " "Ae151"," " "Ae151"," " "Ae151"," "

"Ae151","" "Ae151","" "Ae151","" "Ae151",""

(Hidden Markov Model HMM)

John

saw the man with a telescope. with a telescope the

man() saw(

) PP attachment

12



UNIX LINUX

PC LINUX UNIX

PERL 4

tr

datafile output

tr A-Z a-z < datafile > output

tr sc A-Za-z \012 < datafile > output

sort

ASCII sort datafile > output

sort n datafile > output

sort nr datafile > output

sort +2 nr datafile > output

uniq

uniq datafile > output

uniq c datafile > output

sort datafile | uniq > output

sort datafile | uniq c > output

4
LINUX PERL (Interpreter) PERL
Window

13

tr sc A-Za-z \012 < datafile | sort | uniq c > output

wc

wc l datafile

wc w datafile

wc c datafile

grep

pattern grep pattern datafile > output

awk

pattern

awk /pattern/ { print } datafile > output

datafile tab

awk { print $2 \t $1} datafile > output

datafile 1.65 tab

awk { if ($3 >= 1.65) { print $2 \t $3} } datafile > output

perl C awksedshell programing awk

14



(mutual information):

(mutual information)(information theory)

P ( x, y )
MI ( x, y ) = log 2
P ( x) P ( y )

N f(x,y)

x y f(x)f(y) x y
f ( x, y )
( N 1) N f ( x, y )
MI ( x, y ) = log 2 f ( x) f ( y ) log 2 f ( x) f ( y )

N N

Ken Church (1991)

(word association)5

(collocations)

T-(t-score):

T-T-

5
Using Statistics In Lexical Acquisition.In Zernik, U. (1991) (eds.) Lexical Acquisition: Exploiting
On-Line Resources to Build a Lexion.
15

(t-score)(statistical significance test)

Ken Church (1991)

T-(confidence interval)

T- 1.65 95

T
f ( x) f ( y)
f ( x , y)
t N
f ( x , y)

(entropy):

(entropy)1960 1990

entropy = - kpi log(pi)

N (ngram)(language model)

N N

N-1

(bigram)(trigram)

16



Sketch Engine 1 4

5 Sketch Engine

http://db1x.sinica.edu.tw/kiwi/mkiwi/

17


http://db1x.sinica.edu.tw/kiwi/mkiwi/

()
http://db1x.sinica.edu.tw/kiwi/mkiwi/

18


http://db1x.sinica.edu.tw/kiwi/mkiwi/

http://db1x.sinica.edu.tw/kiwi/mkiwi/

19

 Sketch Engine
http://www.sketchengine.co.uk/

Sketch Engine
http://www.sketchengine.co.uk/

20



Resnik, Olsen, and Diab (1999)

(Open Source)

(1)

21



In recent years calls for democratization of campuses have grown more

insistent. Traditional Chinese concepts of the proper ethical relationship between

students and teachers, in which students accorded teachers the same level of respect

they accorded their own fathers, are dissolving.

(2)

Stories like these are less and less exceptional on university campuses.

Manyprofessors have come to believe that a number of factors have laid hidden bones

of contention in teacher-student relations in recent years politicisation of all aspects of

life, democratisation in society, the promulgation of the new

"University Law" three years ago, rising feminism. . . . Relations have

become much more subtle and complex.

Gao (1998)

(discourse unit)

22



(Brown et al. (1991), Gale and Church (1993) )

(dynamic programming)

Brown et al. (1991) Gale and Church (1993)

(Hansard) 93

Gao (1998)

(Hansard) 89

53 35

Gale and Church (1993)

Category Frequency Prob(match)
1 - 1 1167 0.89
1 - 0 or 0 -1 13 0.0099
2 - 1 or 1 - 2 117 0.089
2 - 2 15 0.011

Bead Types (1,1) (1,2) (1,3) (1,4) (1,5)

frequency 0.53 0.32 0.06 0.06 0.03

23


Bead Types (1, 1) (1,2) (1,3) (1,4) (1,5)
frequency 0.35 0.38 0.17 0.06 0.04

Kay and Roscheisen (1993), Fung and Church (1994)

,,

Gao (1998) Fung Church (1994) Fung Church

(1994) K-vec (mutual information) t (t-score)

(mutual information)

(information theory)

N
f(x,y) x y

f(x)f(y) x y

Log2 ( (P(AB) / (P(A) * P(B) ) = Log2 (f(x,y)/f(x)*f(y))

Ken Church (1991)

(word association)

1.65

T-T-(t-score)

(statistical significance test) Ken Church (1991)

T-
24

(confidence interval) T- 1.65 95

T
P( x| y) P( x| z)
t=
2 P( x| y) + 2 P( x| z)
x|y y x

T N

f(x,y) x y f(x)f(y) x y

f ( x) f ( y)
f ( x, y)
t N
f ( x, y)

Fung Church (1994)

Fung Church K

K K

1 0

Fung Church

5 10 K K

Fung Church (1994)

()
a = k(A B) b = k(~ A B)
c = k(A ~B) d = k(~A ~B)
a b

c d

25

 t

P(Vc) P(Ve)

P( Vc , Ve )
MI( Vc , Ve ) = log2
P( Vc ) P( Ve )
P( Vc ) = a+b
a+b+c+d
P( Ve ) = a+c
a+b+c+d

P( Vc , Ve ) P( Vc ) P( Ve )
t ( Vc , Ve ) =
P( Vc , Ve )
K

Gao (1998)

70 30

Gao (1998) Fung Church (1994)

University of Pennsylvania

Summer Institute of Linguistics

Fung Church

(1994) 90

Fung and Church (1994) K-vec

26

1

2exact string match

teacher

-partial string match

(proximity)

(translation equivalents) 90

Champollion Tool Kit (CTK)

CTK

CTK

Perl

CTK

CTK

27

 CTK

Lucene

28



80 (rule-based)

(data-driven)

(Sinica Chinese Treebank)

Penn Chinese Treebank

(toolkit)Bird,

Klein, and Loper (2009) NLTK

(http://nltk.sourceforge.net/index.php/Main_Page)

(context-free parser)(finite-state

machine) (context-sensitive grammar)

(context-free parser)

Turing Machine

(context-free

parser)(phrase structure rules)

Bird, Klein, and Loper (forthcoming) NLTK python

NLTK NLPNatural language processing

NLP NLTK

NLTK PCFG

(probabilistic context-free parser)

Stanford parser (Klein and Manning (2002))

2002/12/05 parser 2006/6/11 1.5.1

29

Stanford parser PCFG JAVA

parser Dan Klein Christopher Manning

I/O Roger LevyChristopher Manning

Teg GrenagerGalen Andrew Stanford parser

Penn Treebank

Penn Chinese Treebank Penn

Treebank Xue and Xia (2000)

Stanford parser Klein and Manning (2002) factored

parsing maximum

likelihood-estimated PCFG constituent-free dependency parse

Stanford Parser Penn ChineseTreebank

Chiang (2003) Bikel (2004)

(parsing)

context free grammar ambiguous

S VP

S NP

VP V NN

NP V NN

NN

30

 [S [VP [V ] [NN ] ] ]

[S [NP [V ] [NN ] ] ]

CFG parser

(probabilistic context free grammar)

weighted grammar derivation()

PCFG CFG parser

PCFG

derivation

PCFG

Penn Chinese Treebank

NLTK

Viterbi-style Viterbi PCFG parser

(bottom-up parser)(dynamic programming)

most likely constituent

table

1.

2.

3.

4.

31


Most Likely Constituents Table

[0:1] NP (NP ) 0.3

[2:3] NP (NP ) 0.3

[5:6] NP (NP ) 0.3
[1:4] PP (PP (NP ) ) 0.05
[5:6] VP (VP (NP )) 0.03

[0:4] NP (NP (NP ) (PP (NP ) 0.01

))
[0:6] S (S (NP (NP ) (PP (NP ) 0.0001
)) (VP (NP )))

parser S

Viterbi parser PCFG

Viterbi parser 1

234

Inserting tokens into the most likely constituents table...

Insert: |=.....|

Insert: |.=....|

Insert: |..=...|

Insert: |...=..|

Insert: |....=.|

Insert: |.....=|
32

Finding the most likely constituents spanning 1 text elements...

Insert: |=.....| NP -> '' (p=0.15) 0.1500000000

Insert: |.=....| P -> '' (p=0.61) 0.6100000000

Insert: |.=....| NP -> '' (p=0.5) 0.500000000

Insert: |..=...| LC -> '' (p=0.1) 0.5000000000

Insert: |...=..| V -> '' (p=0.61) 0.6500000000

Insert: |....=.| VP -> V (p=0.2) 0.1300000000

Insert: |.....=| NP -> '' (p=0.5) 0.5000000000

Finding the most likely constituents spanning 2 text elements...

Insert: |....==| VP -> V NP (p=0.7) 0.0455000000

Finding the most likely constituents spanning 3 text elements...

Insert: |.===..| PP -> P NP LC (p=1.0) 0.1525000000

Finding the most likely constituents spanning 4 text elements...

Insert: |====..| NP -> NP PP (p=0.25) 0.0057187500

Finding the most likely constituents spanning 5 text elements...

Insert: |.=====| VP -> PP VP (p=0.1) 0.0069387500

Discard: |.=====| VP -> PP VP (p=0.1) 0.0069387500

Finding the most likely constituents spanning 6 text elements...

Insert: |======| S -> NP VP (p=1.0) 0.0002602031

33

 NP Chunking

NP ChunkingNLP

(Ramshaw and Marcus (1995), Kudo and Matsumoto (2000, 2001))

(parsing)semantic role labeling

(co-reference)coherence

(information retrieval) information extractiontext mining

name entity

base NP

(NP conjunction)

94(Kudo and Matsumoto (2000,

2001))

finite state machines pattern (Voutilainen

(1993)) Church (1988),

Chen and Chen (1994)

(Penn Treebank) ( Marcus, Santorini and Marcinkiewicz (1993)),

machine learning

HMM (hidden Markov model)transformation-basedRamshaw and Marcus

1995, memory-based Veenstra 1998, Tjong Kim Sang and Veenstra 1999

Argamon, Dagan and Krymolowski 1998, maximum entropy Skut and Brants

34

1998, SVM (Kudo and Matsumoto, 2000, 2001)

HMM (hidden Markov model) finite state

machine transition function transformation-based

learning transformational rules,

parseHMMtransformation-based learning, memory-based learning

SVM machine learning ,

Wall Street Journal Corpus ,

(precision)(recall) 90%, SVMKudo and Matsumoto

2001(precision)(recall) 94%

Zhao and Huang (1998)

minimum description length principle (MDL)

quasi-dependency strength base NP

unsupervised learning(close test)(open test)

91.5% 88.7%

SVM (treebank)

argument

16

35

 2

NP, VP ,

(:

http://godel.iis.sinica.edu.tw/CKIP/treebank/apposition.htm

terminal node

head(node)

NP, , , , ,

base NP

,,

(2005) Kudo and Matsumoto (2000, 2001)

(I,O,B):

class I,O,B:

I:

O:

B:
36

 Tjong Kim Sang IOB1 Start/End

(Uchimoto et al.2000)
SE I,O,B, class:

B:

E:

I:

S:

O:

Inside/Outside Start/End

I S

O O

I B

I I

I E

B S

, 7 , Wordi i

POSi i :

Wordi POS(i-2) POS(i-1) POSi POS(i+1) POS(i+2)

37

 I 1: 2:0 3:0 4:N 5:S 6:N

O 1: 2:0 3:N 4:S 5:N 6:V

I 1: 2:N 3:S 4:N 5:V 6:N

I 1: 2:S 3:N 4:V 5:N 6:V

I 1: 2:N 3:V 4:N 5:V 6:0

B 1: 2:V 3:N 4:V 5:0 6:0

nominalization

VD(), VK(), VG()

(D) (VK) (Nc) (Na) (P) (Na) (Na)

(VD) (Na)

NP chunks : , ;

NP chunk , SVM ,

( SVM (V)),

VH , NP , V

.,

,
38

 23, 6,

SVM

SVM
Precision Recall
(1) 54.99% 53.17%

(2) 78.18% 59.33%

Kudo and Matsumoto (2000,2001)

(94%) ; :

IOB tag, I/O tag, chunk

( chunk ) tag, IOB

Start/End

SVM

kernel function SVM ,

linear, polynomial, radial basis function, sigmoid... ,

cross validation

: . 8,000

4 , SVM time complexity O(n2),

300,000 , , ,

cross validation , scaling

YAMCHA (http://chasen.org/~taku/software/YamCha/) Taku Kudo NP

Chunking SVM SVM SVM Tool: LIBSVM

39

(Chih-Chung Chang and Chih-Jen Lin, 2004)YAMCHA libsvm

a) Dynamic programming

b) Kernel Function

libsvm , chunking

chunking . , :

Inside/Outside

(B)

SVM tag ,

; IOB tag, (

) , IOB tag ,

YAMCHA Kudo and Matsumoto (2000,2001) IOB tag IO tag ,

B tag NP-chunk , NP-chunk

Kudo and Matsumoto (2000,2001) voting voting

, parsing

(backward ),

parsing SVM , Accuracy

40

. parsing ,

SVM

Kudo and Matsumoto (2000) F measure

F = (2* precision * recall) / (precision + recall) precision recall

recall precision F measure precision recall

YAMCHA Base-NP chunking

parsing
Precision Recall F measure
86.48% 88.41% 87.43%
(Forward) (10360/11980) (10360/11716)
86.29% 85.21% 85.74%
(Backward) (9983/11569) (9983/11716)
87.34% 75.02% 80.71%
(Forward) (8789/10063) (8789/11716)
84.88% 73.84% 78.98%
(Backward) (8651/10192) (8651/11716)
Vote using 88.71% 85.76% 87.21%
Accuracy Rate (10048/11327) (10048/11716)

F measure forward parsing

voting F measure

forward parsing backward

parsing votingrecall

12 16

41



, 10 ,

26 dynamic programming IOB Start/End

95 F measure

(Support Vector Machine)

(dependency parser)

,,

42

 Yamcha

,(Supervised Machine Learning)

, l

(terms), l , O(l)

O(l2 ) l(l1) ,

(feature), R, L, O

, 3(31)

/ 2 = 3 :

()

R ()
O ()
L ()

43

 (Support Vector Machine),,

:,

lLO , lLR , lOR L O,L R, O R ;

n ,

O(l2 (lLO + lLR + lOR ) n)

(binary classifier)

(multiclass classifier)

,,

,,

,,

,,

,(HEADER);,

2008

(Unsupervised Machine Learning)

2008 Google
44



2008

TinySVM YamCha (Kudo Matsumoto (2000))

(Kudo Matsumoto (2002))

(L, R, O)

YamCha

Perl

(2-degree polynomial kernel L1-loss support vector machine),

1(C = 1)( 125 )

()
43253
882708
L O (lLO) 77161
L R (lLR) 34490
O R (lOR) 130692

, 12492 , 246054 ,74850

0.4

45

 l ,,

l , l(l1) (O),

,:

()

57109 (76.298%)
6724 (53.826%)

bank

bank

Senseval-2 English lexical sample 2001 73

(thesaurus) Lesk (1986)

46



Walker (1987)(thesaurus)

(supervised learning)

(unsupervised learning)

Purandare and Pedersen (2004)

Wordnet

(co-occurrence matrix) Singular

Value Decomposition (SVD) 100 Latent Semantic Indexing

(LSI)Jurafsky and Martin (2000)

(collocational features)

bag of words information

Semantic Concordancer Senseval

pseudoword pseudoword Gale et

al. (1992) Schutze(1992)

banana-door banana door

banana-door

duty

Brown et al. (1991) Gale et al.

(1992)

50 (context vector), Bayesian

47

classification

Bayesian classification Bayesian classification

Yarowsky (1995)

(One sense per discourse)

(One sense per collocation)

Yarowsky (1995)Lin (1997)

(classifier)

(knowledge source)

MINIPAR (dependency relations)

(local context)Lin (1997)

Le and Shimazu (2004)

Forward Sequential Selection Algorithm

mutual

information Flip-Flop algorithm (Brown et al. (1991)), decision list

(Yarowsky (1994))

Nave Bayes Maximum EntropySupport Vector Machine

Conditional Random Field

48

 Lucene

Lucene Java

(Inverted File) API Lucene

(open source)Lucene

Lucene

API

Lucene

Lucene

Lucene

Identification of Fields in Texts to

be Indexed : doc (field1, field2 ... )

Indexer

Lucene Index

Searcher

Output of Lucene: Hits (doc

(field1, field2 ...), doc
(field1 ...) ...))

Lucene

49

LDC Sketch Engine

Linguistic Data Constorium (LDC)

Chinese Gigaword

LDC

Sketch Engine

1 2 Sketch Engine 10

50 Sketch Engine

1950

1980

,1990

50



1990 (BNC)

Collin

Collins Cobuild Project

1990

1960 Brown Corpus

IBM AT&T

Google

ETS

ETS

51

 DARPA

NIST

1988

24

HowNet

2. 3.

4.

5.

52





A A Nb N VB Vi



Caa C Nc N VC Vt


Cab POST Ncd N VCL Vt




Cba POST Nd N VD Vt



Cbb C Nep DET VE Vt


D ADV Neqa DET VF Vt


DE , T Neqb POST VG Vt
,
,
Da ADV Nes DET VH Vi



Dfa ADV Neu DET VHC Vt


Dfb ADV Nf M VI Vi



Di ASP Ng POST VJ Vt

53



Dk ADV Nh N VK Vt


FW FW SHI Vt VL Vt


I T T T V_2 Vt

NAV NAV VA Vi



Na , N VAC Vi
,
,

54

 Support Vector Machine (SVM)

SVM machine learning (Boser, Guyon, and Vapnik (1992),

Corts and Vapnik (1995))
, SVM
Joachims 1998 Taira and Haruno 1999 (Kudo and
Matsumoto 2000, 20001),
unknown word
guessing(Nakagawa, Kudo, and Matsumoto (2001)(part of speech
tagging)Nakagawa, Kudo, and Matsumoto (2002) Gimnez J ess and Mrquez
Llus (2004))(dependency analysis)(Kudo and Matsumoto
(2000)) (word sense disambiguation and sense tagging)
(Cabezas, Resnik, and Stevens (2001))(semantic parsing) (Pradhan
et al. (2004) Sun and Jurafsky (2004))
SVM machine,

SVM () ,
, , (
), SVM

55

 min()= (1/2)||2

(, ),
Lagrange multiplier

(), SVM
H : Rd H x
,

.,
T
kernel function: K(xi,xj) = (xi) (xj)
, kernel function .
SVM

(xi, yi) { i= 1, 2, ..., l; xi Rn; yi { 1, -1 }}

l , xi n , yi ( 1
-1) SVM ,

minw,b,e(1/2)wTw + Ci=1lei
yi(wT(xi) + b) >= 1- ei, ei >= 0

xi H SVM
; K(xi,xj) Kernel function.

56

 Bayesian Classification

Bayesian Classification
k
c

57



()
http://ckipsvr.iis.sinica.edu.tw/

AntConchttp://www.antlab.sci.waseda.ac.jp/software.html

Hownethttp://www.keenage.com/

MateParserhttp://barbar.cs.lth.se:8091/

Penn Chinese Treebank

http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2004T05

Sinica Treebank 3.0. http://www.aclclp.org.tw/use_stb_c.php

SketchEnginehttp://www.sketchengine.co.uk/

StanfordNLPSoftwarehttp://nlp.stanford.edu/software/index.shtml

StanfordParserhttp://wwwnlp.stanford.edu/downloads/lexparser.shtml

ToolsforNaturalLanguageAnalysis,GenerationandMachineLearning
http://code.google.com/p/matetools/

YamCha:YetAnotherMultipurposeCHunkAnnotator
http://chasen.org/~taku/software/YamCha/

58


Bikel,Daniel.(2004).OntheParameterSpaceofGenerativeLexicalizedStatistical
ParsingModels.Ph.D.Dissertation.UniversityofPennsylvania.

Bird, Steven, Klein, Ewan and Loper Edward (2009) Natural Language Processing
with Python. OReilley.

Chiang, David. (2003) Statistical parsing with an automatically extracted tree adjoining
grammar. In Data Oriented Parsing, CSLI Publications, pages 299316.

Levy,RogerandManning,ChristopherD.(2003).IsithardertoparseChinese,orthe
ChineseTreebank?.ACL2003.
Manning,Christopher,andSchutze,Hinrich.(1999)FoundationsofStatisticalNatural
LanguageProcessing.MITPress.
Xue,NianwenandXia,Fei.(2000)"The Bracketing Guidelines for the Penn Chinese
Treebank (3.0)",IRCSReport0008,UniversityofPennsylvania,Oct2000.
Argamon,Shlomo,Dagan,Ido,andKrymolowski,Yuval(1998).AMemoryBased
ApproachtoLearningShallowNaturalLanguagePatterns.InProceedingsofthe
17thinternationalconferenceonComputationallinguistics,Vol.1,pp.6773,
Montreal,Quebec,Canada.
Brill, Eric and Ngai, Grace (1999), Man vs. Machine: A Case Study in Base Noun
Phrase Learning. In Proceedings of ACL'99, pp. 65-72, University of Maryland, MD,
USA.
Boser, E. Bernhard, Guyon, Isabelle, and Vapnik, Vladimir. (1992). A Training
Algorithm for Optimal Margin Classifiers. COLT: pp. 144-152
Cabezas, Clara, Resnik, Philip, and Stevens,Jessica. (2001). Supervised Sense Tagging
using Support Vector Machines. Proceedings of the Second International Workshop
on Evaluating Word Sense Disambiguation Systems (SENSEVAL-2), Toulouse,
France, 5-6 July 2001.
Cardie, Claire and Pierce, David (1998). Error-Driven Pruning of Treebank Grammars
for Base Noun Phrase Identification. In Proceedings of COLING-ACL'98, pp.
218-224, Montreal, Canada.
Chang, Chih-Chung and Lin, Chih-Jen. (2004) LIBSVM -- A Library for Support
Vector Machines. http://www.csie.ntu.edu.tw/~cjlin/libsvm/
Chen, Kuang-hua and Chen, Hsin-Hsi (1994). Extracting Noun Phrases from
Large-Scale Texts: A Hybrid Approach and Its Automatic Evaluation, In Proceedings
of ACL-94, Las Cruses, NM, USA.
Church, K. (1988) A Stochastic Parts Program and Noun Phrase Parser for Unrestricted
59

 Text. Second Conference on Applied Natural Language Processing, Austin , Texas ,
pp. 136-143.
Corte, Corinna, and Vapnik, Vladimir (1995). Support-Vector Networks. Machine
Learning 20(3), pp. 273-297.
Gimnez J ess and Mrquez Llus (2004). SVMTool: A general POS tagger generator
based on Support Vector Machines Proceedings of the 4th International Conference
on Language Resources and Evaluation (LREC'04). Lisbon, Portugal. 2004 .
Joachims, Thorsten. (1998) Text Categorization with Support Vector Machines:
Learning with Many Relevant Features. Proceedings of the European Conference on
Machine Learning (ECML), Springer, 1998.
Hsu, Chih-Wei, Chang, Chih-Chung, and Lin, Chih-Jen. (2004). A Practical Guide to
Support Vector Classification.
Kudo, Taku, and Matsumoto, Yuji. (2000). Use of Support Vector Learning for Chunk
Identification. In Proceedings of CoNLL-2000, pp. 142-144.
Kudo, Taku, and Matsumoto, Yuji (2000). Japanese Dependency Analysis Based on
Support Vector Machines, EMNLP/VLC 2000
Kudo, Taku, and Matsumoto, Yuji. (2001). Chunking with Support Vector Machine. In
Proceedings of NAACL 2001, pp. 192-199.
Marcus, Mitchell P., Santorini, Beatrice and Marcinkiewicz, Mary Ann. (1993)
Building a large annotated corpus of English: the Penn Treebank, Computational
Linguistics, 19:2. vol. 19, no. 2, pp. 313330.
Pradhan, Sameer, Ward, Wayne, Hacioglu, Kadri, Martin, James H.and Jurafsky,
Daniel. (2004). Shallow Semantic Parsing Using Support Vector Machines. In
Proceedings of NAACL-HLT 2004,pp. 233-240..
Nakagawa, Tetsuji, Kudo, Taku, and Matsumoto, Yuji. (2001). Unknown Word
Guessing and Part-of-Speech Tagging Using Support Vector Machines. NLPRS, pp.
325-331
Nakagawa, Tetsuji, Kudo, Taku, and Matsumoto, Yuji. (2002). Revision Learning and
its Application to Part-of-Speech Tagging. In Proceedings of ACL 2002, pp.
497-504.
Ramshaw, Lance A., and Marcus, Mitchell P.. (1995). Text Chunking Using
Transformation-based Learning. In Proceedings of the Third ACL Workshop on Very
Large Corpora, pp. 82-94, Cambridge MA, USA.
Skut, Wojciech and Brants, Thorsten. (1998) A Maximum-Entropy Partial Parser for
Unrestricted Text. In Proceedings of the Sixth Workshop on Very Large Corpora, pp.
143-151, Montreal, Canada.
Sun, Honglin and Jurafsky, Daniel. 2004. Shallow Semantic Parsing of Chinese. In
Proceedings of NAACL-HLT 2004, pp.192-199.

60

Taira, Hirotoshi, Haruno, Masahiko 1999 : Feature Selection in SVM Text
Categorization. AAAI/IAAI 1999, pp. 480-486.
Tjong Kim Sang, Erik F. and Veenstra, Jorn (1999). Representing Text Chunks. In
Proceedings of EACL'99, 173-179, Bergen, Norway.
Tjong Kim Sang, Erik F. (2002) Memory-Based Shallow Parsing. Journal of Machine
Learning Research, Vol. 2, pp. 559-594.
Uchimoto,Kiyotaka,Ma,Qing,Murata,Masaki,Ozaku,Hiromi,Isahara,Hitoshi.(2000)
Namedentityextractionbasedonamaximumentropymodelandtransformation
rules.Proceedingsofthe38thAnnualMeetingonAssociationforComputational
Linguistics,HongKong,pp.326335.
Veenstra, Jorn. (1998). Fast NP chunking using memory-based learning techniques, In
F. Verdenius and W. van den Broek eds.,Proceedings of BENELEARN-98, pp. 71-79,
Wageningen, The Netherlands.
Voutilainen,A.(1993) NPtool,aDetectorofEnglishNounPhrase.InProceedingsof
theFirstAnnualWorkshoponVeryLargeCorpora,pp.4857.BernersLee,Tim.
(2000)WeavingtheWeb:theoriginaldesignandultimatedestinyoftheWorld
WideWebbyitsinventor.NewYork:HarperBusiness.
Boguraev,Branimir.andBriscoe,Ted.(1989)ComputationalLexicographyforNatural
LanguageProcessing.Longman:Harlow.Boguraev,BranimirandPustejovsky,James
(eds.)(1996)CorpusProcessingforLexicalAcquisition,MITPress.
Chaffin,RogerandIllerrmann,Douglas.(1988)TheNatureofSemanticRelations:a
ComparisonsofTwoApproaches.InEvens(eds)(1988),pp.289334.
Church,K.andHanks,P.(1990)WordAssociationNorms,MutualInformation,and
Lexicography.ComputationalLinguistics,Vol.16,No.1,pp.2229.
Church,K.etal.(1991)Parsing,WordAssociations,andTypical
PredicateArgumentRelations.InTomita(ed)CurrentIssuesinParsingTechnology,
Kluwer.
Church,Kenneth,WilliamGale,PatrickHanks,andDonaldHindle.(1994)Lexical
Substitutability,inAtkinsandZampolli(eds.)ComputationalApproachestothe
Lexicon,pp.153177.Oxford, OxfordUniversityPress.
Cruse,Allan.(1986)LexicalSemantics.Cambridge:CambridgeUniversityPress.
Dong,ZhendongandDong,Qiang.(2006)HownetandtheComputationofMeaning.
WorldScientific.
Evens,Martha.(eds.)(1988)RelationalModelsoftheLexicon:Representing
KnowledgeinSemanticNetworks.CambridgeUniversityPress.
Fillmore,Charles.(1968)TheCaseforCase.InE.BachandR.T.Harms,eds.,Universals
inLinguisticTheory,Holt,RiinehartandWinston,NewYork,188.
Koenig,JeanPierre.(1999)LexicalRelations.CSLI,StanfordUniversity.

61

Girju,R.,Nakov,P.,Nastase,V.,Szpakowicz,S.,Turney,P.,andYuret,D.(2007),
SemEval2007Task04:ClassificationofSemanticRelationsbetweenNominals,
ProceedingsoftheFourthInternationalWorkshoponSemanticEvaluations
(SemEval2007),Prague,CzechRepublic,pp.1318.
Grefefenstette,Gregory.(1994)ExplorationsinAutomaticThesaurusDiscovery.
KluwerAcademicPublishers.
Hearst,M.A.(1992).Automaticacquisitionofhyponymsfromlargetextcorpora.In
ProceedingsoftheFourteenthInternationalConferenceonComputational
Linguistics,pages539545,Nantes,France.
Levin,Beth.(1985)Introduction,inB.Levin(ed.)LexicalSemanticsinReview,Lexicon
ProjectWorkingPapers1,CenterforCognitiveScience,MIT,pp.162.
Melcuk,Igor.(1988)TheExplanatoryCombinatoryDictionary,inM.Evens(ed.)
(1988),pp.4174.
Pustejovsky,James,SabineBergler,andPeterAnnick(1993)LexicalSemantic
TechniquesforCorpusAnalysis,ComputationalLinguistics,Vol.19,No.2,pp.331
358.
Pustevojsky,James.(1995)TheGenerativeLexicon.TheMITPress.
Pustevojsky,James.(2000)SyntagmaticProcesses.inHandbookofLexicologyand
Lexicography,deGruyter,2000.
Jackendoff,Ray.(1983)SemanticsandCognition.Cambridge,Mass.:MITPress.
Jackendoff,Ray.(1990)SemanticStructures.Cambridge,Mass.:MITPress.Jones,
Stevens.(2002).Antonymy:ACorpusbasedPerspective.London;NewYork:
Routledge,2002

Pedersen,Patwardhan,andMichelizzi(2004)WordNet::SimilarityMeasuringthe
RelatednessofConceptsAppearsintheProceedingsoftheNineteenthNational
ConferenceonArtificialIntelligence(AAAI04),pp.10241025,July2529,2004,
SanJose,CA(IntelligentSystemsDemonstration)

Resnik,Phillip.(1992)WordNetandDistributionalAnalysis:AClassbasedApproach
toLexicalDiscovery,inWorkshopNotes,StatisticallyBasedNLPTechniques,
AmericanAssociationforArtificialIntelligence,pp.109113.
Schank,Roger.(1975)ConceptualInformationProcessing.Amsterdam:NorthHolland.
Sinclair,John.(eds).(1987)Lookingup.Glasglow:Collins.
Turney,P.D.(2006),Expressingimplicitsemanticrelationswithoutsupervision,
Proceedingsofthe21stInternationalConferenceonComputationalLinguisticsand
44thAnnualMeetingoftheAssociationforComputationalLinguistics
(Coling/ACL06),Sydney,Australia,pp.313320.
Wilks,A.Yorick(1968)OnlineSemanticAnalysisofEnglishTexts.MachineTranslation,

62

 Vol. 11, pp. 5972. Brown, Peter et al. (1991) Word sense disambiguation using
statisticalmethods.InACL29,pp.264270.
Dong,ZhendongandDong,Qiang.(2006)HownetandtheComputationofMeaning.
WorldScientific.
Gale, William, Church, Kenneth, and Yarowsky, David. (1992) A method of
disambiguating word senses in a large corpus. Computers and the Humanties
26:415439.
Jurafsky, Daniel, and James H. Martin. (2000) Speech and Language Processing: An
Introduction to Natural Language Processing, Speech Recognition, and
ComputationalLinguistics.PrenticeHall.
Klein, Dan. and Manning, Christopher. (2003) Accurate Unlexicalized Parsing.
Proceedings of the 41st Meeting of the Association for Computational Linguistics,
pp.423430.

Le,CuongAnhandShimazu,Akira.(2004)HighWSDAccuracyUsingNaveBayesian
ClassifierwithRichFeatures.PACLIC18,Tokyo.
http://dspace.wul.waseda.ac.jp/dspace/bitstream/2065/564/1/oral8.pdf

Lesk,Michael.(1986)AutomaticSenseDisambiguation:Howtotellapineconefrom
anicecreamcone.InProceedingsofthe1986SIGDOCConference,pp.2426,
NewYork.AssociationforComputingMachinery.

Lin,Dekang.(1997).UsingSyntacticDependencyasLocalContexttoResolveWord
SenseAmbiguityInProceedingsofACL97,Madrid,Spain.July,1997.
Manning,Christopher,andSchutze,Hinrich.(1999)FoundationsofStatisticalNatural
LanguageProcessing.MITPress.
Patwardhan,Banerjee,andPedersen(2005)SenseRelate::TargetWordAGeneralized
FrameworkforWordSenseDisambiguation.AppearsintheProceedingsofthe
TwentiethNationalConferenceonArtificialIntelligence,July12,2005,Pittsburgh,
PA.(IntelligentSystemsDemonstration)
PurandareandPedersen(2004)ImprovingWordSenseDiscriminationwithGloss
AugmentedFeatureVectors.AppearsintheProceedingsoftheWorkshoponLexical
ResourcesfortheWebandWordSenseDisambiguation,November22,2004,
PueblaMexico.
Yarowsky,D.(1994)DecisionListsforLexicalAmbiguityResolution:Applicationto
AccentRestorationinSpanishandFrench.''InProceedingsofthe32ndAnnual
MeetingoftheAssociationforComputationalLinguistics.LasCruces,NM,pp.
8895.
Zhao,JunandHuang,Changning.(1998).AQuasiDependencyModelforStructural
AnalysisofChineseBaseNPs.InProceedingsofCOLINGACL98,pp.17,Montreal,
63

Canada.

(2007)
,pp.pp257272

(2007) :
,pp131144

2005
pp.317332

2005
pp.385396.

2004

(2001)

(Sinica Treebank Version 3.0).

http://www.aclclp.org.tw/use_stb_c.php

(1988). ,

64