A System of Verbal Semantic Attributes Focused on 
the Syntactic Correspondence between Japanese and English 
Hiromi NAKAIWA, Akio YOKOO and Satoru IKEHARA 
NTI" Communication Science Laboratories 
1-2356 Take, Yokosuka-shi, Kanagawa-ken, 238-03 Japan 
E-mail: nakaiwa {yokoo, ikehara} @nttkb.ntt.jp 
Abstract 
This paper proposes a system of 97 verbal 
semantic attributes for Japanese verbs which 
considers both dynamic characteristics and the 
relationship of verbs to cases. These attribute 
values are used to disambiguate the meanings of 
all Japanese and English pattern pairs in a 
Japanese to English transfer pattern dictionary 
consisting of 15,000 pairs of Japanese valence 
patterns and equivalent English syntactic 
structures. 
1. Introduction 
Various machine translation systems have 
approached the stage of being put to practical 
use. However, the quality of the finished 
translation is not satisfactory in any of these 
systems. This is due to difficulties in limiting 
linguistic phenomena that are handled by 
machine translation systems. In particular, the 
analysis of linguistic expressions such as ellipsis 
and anaphoric references, which require 
contextual analysis, is imperfect. To introduce 
constraints brought about by context requires an 
enormous volume of knowledge of word 
meanings that can be used to determine the 
semantic relationships between one sentence and 
another. 
To avoid an explosion in the volume of 
knowledge, a technique is proposed for 
classifying word meanings and determining the 
relationships between words or between 
sentences using the typical attribute values of 
each word. Particularly in the case of context 
processing, the verbal semantic attributes that 
become the key factors in analyzing the flow of 
sentences constitute important knowledge. 
Various efforts have been made in 
researching verb classification. Muraki (1985) 
suggested a method for grouping Japanese verbs 
using their word meanings and their syntactic 
features. Tomiura et al. (1986) proposed a 
method for representing the meaning of verbs 
divided into fundamental meanings and 
reasoning rules. Ogino et al. (1989; EDR 1990) 
proposed a method for verb classification based 
on relations between verbs and co-occurring 
elements. Various efforts have been made to 
classify English verbs. For example, Levin 
(1993) proposed a method for the classification 
of 3000 English verbs that uses the relationship 
between syntactic behavior and shared meaning. 
The research about verb classification still 
tends to be limited solely to classification of the 
semantics of verbs per se. It does not take into 
account the relationship between word meanings 
and their usage within sentences and is not 
aimed at natural language processing. Thus, the 
full benefits that could be achieved in the 
analysis of tracking semantic relationships 
between sentences and eliminating the polyscmy 
of verbs have not been realized. 
This paper tbcusses on the relationship 
between word meanings of verbs and their 
usage, and seeks to classify the semantic 
attributes of verbs. These semantic attributes are 
used in defining the method of use of each verb 
in Japanese to English transfer pattern 
dictionaries. They furnish the key to tracing the 
semantic relationships of verbs that are used in a 
text. 
2. Semantic Structure of Verbal 
Patterns 
This chapter examines the relationship between 
the usage of verbs I and the semantic structure of 
verbs. In machine translation systems, it is well 
known that the translation pattern pairs of source 
IALT-J/E's pattern dictionaries include both verbs and 
adjectives. Japanese adjectives are the equivalent of 
English 'be Adjective': for example "A-ga utsukushii" => 
"A is beautiful". 'verbs' will be used to refer to both verbs 
and adjectives from now on. 
672 
hmguage and target language sentence are 
effective in detmrnining the meaning of verbs. 
Our machine translation system, AI,T-J/E, 
uses two types of Japanese to English transfer 
pattern dictionaries based on verbs: the semantic 
valence pattern transfer dictionary and the 
idiomatic expression transfer dictionary (Fig. 1). 
These dictionaries consist of pairs made of 
\[The semantic valence Imttern transfer d ictimmryl 
(1) NI(SUBJECI'S)-ga N2(F(X)D)-wo taheru 
eat 
=> N1 eatt N2 
(2) Nl(*)-ga yomigaeru 
revive 
:> NI revive 
\[The idiomatic expression transfer dictionary l 
(1) NI(sUBJECTS)-ha se-ga takai 
back high 
=> NI is tall. 
Fig. 1 Japanese to English Transfer Pattern 
Dictionaries 
(The semantic constraints are shown in parenthesis, 
* indicates flmre is no senmntic constraint.) 
Japanese unit sentence patterns derived from 
Japanese verbs 2 with semantic constraints to 
their case elements and English patterns which 
correspond to thc Japanese expressions, t"or 
example, pattern(l) in Fig. 1 shows how, if the 
Japanese verb is "taberu" and the noun phrase 
with a "ga" particle, which shows a subject, has 
the semantic attribute SUBJECTS and the noun 
phrase with a "we" particle, which shows a 
direct object, has the setnantic attribute FOOl) 
then the verb should be translated as "eat". The 
noun phrase with the "ga" particle is translated 
as the English subject. The noun phrase with the 
"we" particle is translated as the English direct 
object. Here, wc exantine the rehttionship 
between the usage of verbs and the semantic 
structure of verbs using verbal patterns that have 
been entered into the Japanese to English 
transfer pattern dictionaries of ALT-J/IL 
Fig. 2 shows an example of entries in 
the Japanese to English transfer pattern 
dictionary which indicate the patterns of the 
2In the idiomatic expression transfer dictionzu'ies, these arc 
the core secg)r of idiomatic exp,essions such as "Abura we 
uru" literally, "to sell oil", but kliomatically, "to idle away 
time". 
Japanese verb "tsutsumu". This verb has three 
patterns. 
\[Japanese Verb : tsutsumu "wrap"l 
(l) NI(SUBJECTS)-ga 
N2(CONCRETE OBJECTS or PEOPLE)-wo 
N3(CLOTHES or PAPERS)-de 
IsuIsumu 
=> N1 wrap N2 in/withN3 
Verbal Semantic Attribute: NI's bodily action 
(2) NI(FIRE, ATMOSI'tlERI~ or AlR)-ga 
N2(CONCRETE OBJECTS, CULTtJRE or PLACES)-wo 
Isulsumu 
=> Ni envelop N2 
Verbal Semantic Attribute: N1 clmnges N2's attributes 
(3)NI(FOG)-ga 
N2(CONCRETE OBJECTS or PI,ACES)-wo 
ISI~tsIImu 
=> NI veil N2 
Verbal Semanlic Attribute: Natural Plmnomena 
Fig. 2 Example of a Japanese Verb with 
multiple patterns 
The first example shows a pattern pair 
indicating that the equiwtlent of the Japanese 
expression "N 1 (SUBJECTS) ga N2 (CONCRETE 
OBJECTS or PEOPI,E) we N3(CLOTHtiS or I'APERS) 
de tsutsumu" is the English expression "N1 wrap 
N2 in/with N3". When the Japanese verb 
"tsutsumu" was used with these cases, this 
sentence gives the impression that NI really 
does the wrapping action. So, in this case, this 
pattern has the w:rb meaning "N1 conducts 
bcxlily action.". 
The second example shows a pattern pair 
indicating that the equivalent expression of the 
Japanese expression "NI (FIRE, ATMOSPIIERE or 
AIR) ga N2(CONCRETE OBJECTS, CULTtlRE or 
PLACES) we laulsumu" is the English expression 
"NI envelop N2". This sentence gives the 
impression that the state of N2 which isn't 
usually enveloped by N1 changes to the 
enveloped state. So, even though the same 
Japanese verb "ISUL~'UmU" was used with these 
cases, in this case, the pattern has a verb 
meaning of "NI changes N2's attributes.". 
The third example shows a pattern pair 
indicating that the equiwdent of the Japanese 
expression "N 1 (FOG) g(l N 2(C()NCRETF~ OBJECI'S 
or PLACES)we tsutxumu" is the English 
expression "N1 veil N2". In this case, this 
sentence gives the impression that a natural 
6Z3 
phenomenon, 'fog', has occurred. So, this 
pattern has the meaning "Natural Phenomena 
have arisen". 
As shown in these examples, maintaining 
expressions in pairs which indicate both the 
common meaning and their usage between the 
Japanese and English, enables us to eliminate 
many conceptual ambiguities and makes it 
possible to give detailed and accurate attribute 
values to the Japanese verb "tsutsumu". 
As in the case of the Japanese verb 
"tsutsumu", one verb normally has several kinds 
of conceptual structures. But one verbal pattern 
which indicates common word meanings and 
their use between the Japanese and English 
(which differ so vastly in syntactic structure) 
corresponds to one conceptual structure. So, it is 
possible to eliminate the conceptual ambiguity 
of verbs by selecting verbal patterns in syntactic 
semantic analysis. In Japanese to English 
machine translation, we estimate there are tens 
of thousands of verbal patterns which need to be 
defined. If the usage of these patterns can be 
expressed by a small number of verbal semantic 
attributes, it is possible to track the semantic 
relationships of verbs easily. When giving 
verbal semantic attributes to a pair of individual 
Japanese and English patterns, it is possible to 
refer to the meaning of verbs not only in 
Japanese but also in English. 
3. System of Verbal Semantic 
Attributes 
3.1 Classification Standards for 
Verbal Semantic Attributes 
Regarding the classification of verbs for use in 
machine translation, Nishida et al. (1980) 
proposed a system of verbal classification. This 
system of classification was introduced to 
resolve syntactic and semantic ambiguities of 
English in English to Japanese machine 
translation. To this system, they added the 
semantic attributes of verbs to the patterns of 
English verbs proposed by Hornby (1975) and 
determined the case structures depending on the 
combination of these two kinds of information. 
This system of verbal semantic attributes was 
introduced on the condition that the features of 
syntactic structures are expressed by Hornby's 
patterns of English verbs. So, this system of 
classification focused only on word meaning. 
Therefore this system can not be applied as such 
to the classification of Japanese verbs because 
Hornby's patterns can't be applied directly to 
Japanese verbs. No one has yet to propose 
exhaustive patterns like Hornby's for Japanese 
verbs. 
We expanded our system based on the 
discussions in section 2, using the following two 
factors. 
• Dynamic Characteristics of verbs 
Classification based on a verb's meaning 
and its effects on the discourse: 
This classification is based on the types of 
action that can be understc×)d to have occurred 
when a verb is expressed and what situations 
have been brought about. 
Ex. "motsu"(to have) -- Possession 
"kaihatsusuru"(to develop) -- Production 
The verb "motsu" indicates that there is an 
act of possession within the context. In contrast, 
the verb "kaihatsusuru" indicates that there is 
something being produced within the context. 
• Relationslfip of Verbs to Cases 
Classification based on the role which the 
cases play with the verbs that govern them: 
This classification is based on the roles played 
by the case elements governed by the verb 
expressed. 
Ex."kanseisuru":SUBJ be completed 
->SUBJ be produced 
"kaihatsusuru":SUBJ develop OBJ 
->SUBJ produce OBJ 
"kanseisuru" and "kaihatsusuru" "are both 
verbs which indicate acts of production. But 
whereas "kanseisuru" indicates that the SUBJ is 
being produced, "kaihatsusuru" indicates that the 
SUBJ produces the OBJ. 
3.2 Semantic Attribute System considering 
the Semantic Relationship between Verbs 
We created a system of verbal semantic 
attributes as explained above. The semantic 
attribute values were determined using the 
usage patterns of typical Japanese verbs. First 
we classified verbs focussing on their dynamic 
characteristics. Next, we classified each group 
again focussing on the relationships of verbs to 
674 
their cases. The top levels of the created system 
of verbal semantic attributes are shown in Fig. 3. 
The left side of this figure lists classifications as 
based on the dynamic characteristics of the verbs 
(their meanings). The right side lists the 
classifications based on the relationship of verbs 
with their cases (their usage). On the basis of 
these classification criteria, 97 verbal semantic 
attributes have been established. 
EVENT 
Dynamic Characteristics of Verbs 
STATE 
ACTI()N~ 
Abstract 
-- Relation-- 
__ Menral 
State 
-- Natnre 
__ Physic~d 
Action 
Existence 
Attribute 
-- Possession 
-- Relative 
Relation 
__ Relatiou of Cause 
and F, ffcct 
___~ Perceptual State 
Emotive State 
Thinking Stale 
-- Physical Transfer 
-- Possessive 'l'rans\[~: 
-- Attribute Transl~r - 
__ Bodily Transfer 
Result 
Bodily Action 
-- Use 
-- Connective Action- 
- Prod uction 
__ I,;xtinction- 
Destruction 
Mental F Mental Transfer 
Action "-1 Become \[....._ Perceptual 
1 Action Cause 
1---- Emotive Action Enable 
Start-End---~ Sutrt /L.__ Thinking Action End 
4. Result of Applicationfor the Semantic 
Descriptions of Verbal Patterns 
We evaluated the coverage of the verbal 
semantic attributes shown in chapter 3 by 
examining the verbal semantic attributes for 
each Japanese to English pair (about 15,000 
pairs) in the Japanese to English transfer pattern 
dictionaries 3. 
Fig.4 shows how many transfer patterns 
were created for each verb in the semantic 
Relationship between 
Verbs and Cases 
-- SUBJ exist 
SUBJ not exist 
Relation between 
SUBJ and DIR-OBJ 
Relation between 
SUFIJ and IND-OBJ 
SUBJ cause INI)-OBJ 
SUBJ cause I)IR-OBJ 
tc,; 
iiiiiiiiiiii 
.~ SUBJ be accepted 
S.U!~.J..p. rovidcs IND-OI~J 
Z :::::::::::: 
f: U: 
E 
C SUBJ belmxluced 
SUBJ pr(xluce OBJ 
C :::::::::::: 
Fig. 3 System of Verbal Semantic Attributes 
valence pattern transfer 
dictionary and the 
idiomatic expression 
transfer dictionary. This 
figure shows the results 
that were counted for 
each different verb. "File 
percentage of patterns 
that came from verbs 
with more than one 
pattern was 73.4%. In 
these verbs that have 
multiple patterns, the 
percentage that had 
different kinds of verbal 
semantic attributes added 
to the patterns were 
70.1%. This result 
shows that it is possible 
to classify semantic 
attributes for each verb 
by adding verbal 
semantic attributes to 
Japanese and English 
transfer pairings. 
Next we counted 
the number of verbal 
semantic attribute values 
givett for each pattern. 
Fig. 5 shows how many 
verbal semantic attributes 
3Attribute values from a 
general noun attribute system 
classified into some 2,800 
types have been i)rovidcd as 
selnantic constraints to the 
case elements of these patterns 
(lkchara ct al. 1991) enabling 
accurate selections of 
syntactic structures. 
675 
la0 
45 
40 
35 
30 
20 
15 
10 
5 
| ~ i t • i i • 0 
O3 
Number of Patterns 
Fig. 4 Ratio of the number of patterns to 
each verb 
were used by how many patterns. About 90% of 
patterns can be described by just one attribute 
value. This result shows that by giving the 
verbal semantic attributes proposed in this paper 
to each pattern in ALT-J/E, even in instances 
where multiple meanings may exist for a given 
Japanese verb, meanings can be selectively 
limited when verbs are viewed in terms of 
pattern pairings. The verbal semantic attributes 
which were given in each pattern have the 
potential to become an important key to tracking 
semantic relationships between sentences as is 
shown in chapter 5. 
Fig.6 shows the most frequent ten verbal 
semantic attributes for all the patterns. In these 
verbal semantic attributes, the patterns that 
ATTRIBUTE was added can almost all be 
described by only one attribute value (26.4% out 
of 27%). By contrast, the many patterns 
100 
90 
8O 
70 
60 # 
so 
40 m 
30 
20 
10 
1 2 3 4 5 
Number of Added Verbal Semantic Attributes 
Fig. 5 Ratio of the number of added verbal 
semantic attributes to each pattern 
No. 1 :ATTRIBUTE, Coverage: 27.0% 
Number of added VSA: 1:26.4%, 2 or more:0.6% 
Ex. NI(SUBJECTS)oga chikarazuyoi 
=> N1 be reliable 
No.2 :BODILY ACTION, Coverage: 12.7% 
Number of added VSA: 1:9.9%, 2 or more:2.8% 
Ex. NI(HUMAN)-ga odoru 
=> N1 dance 
No.3 :ATTRIBUTE TRANSFER(Subj's attribute), 
Coverage: 9.4% 
Number of added VSA: 1:8.1%, 2 or more: 1.3 % 
Ex. NI(CONCRETE OBJECTS)-ga 
N14(FIRE, HEAT or LIGHT)-de tokeru 
=> N1 be melted by N14 
No.4 :THINKING ACTION, Coverage: 8.9% 
Number of added VSA: 1:7.5%, 2 or more: 1.4% 
Ex. NI(HUMAN)-ga N2(CULTURE)-wo 
fukushuusuru 
=> N1 review N2 
No.5 :ATTRIBUTE TRANSFER 
(Subj changes Dir-Obj's attribute) ,Coverage: 7.9% 
Number of added VSA: 1:5.8%, 2 or more:2.1% 
Ex. NI(SUBJECTS)-ga 
N2(PRODUCTS or CULTURE) wo 
moyasu 
=> N1 burn N2 
No.6 :EMOTIVE ACTION(Subj acts), Coverage: 7.7% 
Number of added VSA: 1:6.6%, 2 or more: 1.1% 
Ex. NI(SUBJECTS)-ga N2(DEATH)-wo 
kanashimu 
=> N1 mourn N2 
No.7 :MENTAL TRANSFER 
(Subj transfers Dir-Obj to Ind-Obj), Coverage: 4.9% 
Number of added VSA: 1:4.5%, 2 or more:0.4% 
Ex. NI(SUBJECTS)-ga 
N2(LITERATURE)-wo 
N3(PUBLICATION or BOOK)-ni 
kankousuru 
=> N1 publish N2 in N3 
No.8 :EMOTIVE STATE, Coverage: 2.1% 
Number of added VSA: 1:1.8%, 2 or more:0.3% 
Ex. NI(SUBJECfS)-ha N2(ABSTRACT)-ga 
kuyashii 
=> NI regret N2 
No.9 :RELATIVE RELATION 
(between Subj and Ind-Obj), Coverage: 1.8% 
Number of addcA VSA: 1:1.3%, 2 or more:0.5% 
Ex. NI(HUMAN)-ga N2(CULTURE)-wo 
N3(ttUMAN)-ni shijisuru 
=> NI study N2 under N3 
No. 10:POSSESSIVE TRANSFER 
(Subj provides Ind-Obj), Coverage: 1.6% 
Number of added VSA: 1:1.4%, 2 or more:0.2% 
Ex. NI(SUBJECI'S)-ga N3(SUBJECFS)-ni 
zouwaisuru 
=> NI bribe N3 
Fig. 6 Coverage of the top 10 verbal 
semantic attributes 
676 
described by BODILY ACTION or ATTRIBUTE 
TRANSFER was added can't be described by one 
attribute value (2.8% out of 12.7% and 2.1% out 
of 7.9 %, respectively). These 2 kinds of 
attribute values indicate the SUBJECT'S Physical 
Action, and it tends to be difficult to resolve the 
semantic ambiguities for these patterns. 
As shown in Fig.6, a few verbal semantic 
attributes cover a large proportion of patterns. 
For example, the sum of the coverage of the 
most frequent attribute value, A'KFRIBUTE, and 
the second most fi'equent attribute value, BODILY 
ACTION, cover 39.7 % of all patterns. For these 
attributes, even if there are several patterns for a 
given verb, sometimes the same attribute value 
was given to all the patterns. So the system of 
verbal semantic attributes is not sufficient to 
resolve the semantic ambiguities. For such 
attributes, we need more detail. We are plalming 
to subdivide these attribute values in the future. 
5. Applications for Context Processing 
in this chapter, we show examples of 
applications in context processing. 
5.1 Analysis of Anaphoric Reference of 
Japanese Zero Pronouns 
Using verbal semantic attributes to analyze 
anaphoric referents of zero pronouns appearing 
in Japanese texts is one applicati(m that has been 
considered (Nakaiwa et al. 1992). This 
technique pays attention to verbal semantic 
attributes and the relationship between the 
semantic attributes of tim verbs which govern 
zero pronouns and the semantic attributes of ttle 
verbs which govern case element candidates 
which may be anaphoricatly referred to. The 
contexts are carefully examined to determine 
anaphoric reference elements. 
This method has been realized in the 
machine translation systmn AIA'-J/E. The 
enhanced ALTJ/E was assessed by processing 
common Japanese newspaper articles. It was 
found that 95% of the Japanesc zero pronouns 
requiring anaphoral resolution in the 102 
sentences from 30 newspapcr articles' lead 
paragraphs tlad their referents determined 
correctly using rules tuned for the 102 
sentences(window test). In tile case of a blind 
test, the rate of success in anaphora resolution in 
which the zero pronoun referent exists within the 
sentence in another 98 sentences from 
newspaper articles was about 83% using tile 
rules. To demonstrate the effectiveness of this 
method, we evaluated the performance of the 
method proposed by Walker et.al. (1990) using 
the 98 sentences. Its rate of success in anaphora 
resolution where the zero pronoun referents 
existed within the sentence was about 74%. This 
result shows that our method is more effective 
than Walker's method, and that the rules used in 
our method determine universal relationships 
between verbs. If a few rules appropriate ff)r tile 
98 sentences are added, tile rate increases to 
95%. This result shows that the load imposed by 
rule customization is low. 
Even in the case of sentences in machine 
translation systems for which target meas cannot 
be constrained, this method allows the 
construction of rules independent of the 
translation target areas by means of verbal 
semantic attribute pairings. Using the verbal 
semantic attributes, anaphoric reference 
resolution of zero pronouns can be conducted 
with a limited volume of knowledge. 
5.2 Supplementation of Elements Outside 
Sentences against Elliptical Case Elements 
Verbal semantic attributes can be used with 
elliptical case elelnents in Japanese texts to 
supplement case elements whose referents do 
not appear within tim texts. To analyze such 
elliptical phenomena, it is possible to use case 
elements' semantic constraint conditions to 
estimate supplementary elements. Semantic 
information used to estimate supplementing 
elelnents is a constraint on cases for selecting the 
transfer f)attcrn. With this xnethod, therelbre, the 
majority of the constraints involve abstract 
semantic information, fi'equently posing 
difficulties in pinpointing elements to be 
supplemented. For example, if in Fig. 1(2), 
"Ni(*)--ga yomigaeru(revive)", N l were to be 
omitted, ttle case element N I has no seman|ic 
constraint, and supplementary elements to the 
case can't be determined. In this case, it is 
effective Io complete the case element 
corresponding to S|JBJECT using tilt" verbal 
semantic attributes of the pattern, "N i's b(×tily 
state is transfcrled". Thus if a method 
presuming supplementary elements of elliptical 
677 
case elements corresponding to the verbal 
semantic attributes is used, the deduction of 
more accurate supplementary elements would be 
possible. 
5.3 Application for Other Context Processings 
The verbal semantic attributes can be applied to 
other context processing problems. Estimating 
the relationship between verbs by pairing of the 
verbal semantic attributes, analysis of the tenses 
relationship of events as indicated by certain 
sentences and events indicated by another, 
together with sentence abridgment can be 
considered. 
6. Conclusion 
This paper has proposed a system of 97 verbal 
semantic attributes for Japanese verbs which 
considers dynamic characteristics and the 
relationship of verbs to cases. These attribute 
values were used to disambiguate the meanings 
of all Japanese and English pattern pairs in a 
Japanese to English transfer pattern dictionary 
consisting of 15,000 pairs of Japanese valence 
patterns and equivalent English syntactic 
structures. As a result of examining the verbal 
semantic attributes for each pattern of Japanese 
to English paring, 90% of patterns can be 
described by only one attribute values. This 
result shows that the meanings of Japanese verbs 
determined by the verbal semantic attributes can 
be effectively limited when verbs are viewed in 
terms of pattern parings. Further attentions to 
details and tightening of standards together with 
extensive application of this system are now 
being worked on. 
References 
EDR (1990) Concept Dictionary, TR-027. 
Hornby, A. S. (1975) Guide to patterns and 
usage in English, 2nd edition, London, 
Oxford University Press. 
Ikehara, S., M. Miyazaki and A. Yokoo (1991) 
Semantic Analysis Dictionary for Machine 
Translation. Information Processing Society 
of Japan, Natural Language Processing, 
Vol.84-13 (in Japanese). 
Levin, B. (1993) English Verb Classes and 
Alternations, The University of Chicago 
Press. 
Muraki, S. (1985) Jyutsugo-so niyoru dousi no 
bunrui (Classification of Verbs by 
Predicates). Information Processing Society 
of Japan, Natural Language Processing, 
Vol.48-5 (in Japanese). 
Nakaiwa, H and S. Ikehara (1992) Zero Pronoun 
Resolution in a Japanese to English Machine 
Translation System by using Verbal Semantic 
Attributes. Proc. of ANLP '92, pp. 201-208. 
Nishida, F. and S. Takamatsu (1980) English- 
Japanese Translation through Case-Structure 
Conversion. Proc. of COLING '80, pp. 447- 
454. 
Ogino, T. et al. (1989) Verb Classification Based 
on Semantic Relation of Co-occurring 
Elements. Information Processing Society of 
Japan, Natural Language Processing, Vol.7l- 
2 (in Japanese). 
Tomiura, Y. and S. Yoshida. (1986) A Research 
of the Polysemy and Description of Verbs. 
Information Processing Society of Japan, 
Natural Language Processing, Vol.55-2 (in 
Japanese). 
Walker, M., M. lida and S. Cote (1990) 
Centering in Japanese Discourse. Proc. of" 
COLING'90. 
678 
