Automatic Acquisition of Hyponyms 
~om Large Text Corpora 
Marti A. Hearst 
Computer Science Division, 571 Evans Hall 
University of California, Berkeley 
Berkeley, CA 94720 
and 
Xerox Palo Alto Research Center 
marti~cs, berkeley, edu 
Abstract 
We describe a method for the automatic acquisition 
of the hyponymy lexical relation from unrestricted 
text. Two goals motivate the approach: (i) avoid- 
ance of the need for pre-encoded knowledge and (ii) 
applicability across a wide range of text. We identify 
a set of lexico-syntactic patterns that are easily rec- 
ognizable, that occur iYequently and across text genre 
boundaries, and that indisputably indicate the lexical 
relation of interest. We describe a method for discov- 
ering these patterns and suggest that other lexical 
relations will also be acquirable in this way. A subset 
of the acquisition algorithm is implemented and the 
results are used to attgment and critique the struc- 
ture of a large hand-built thesaurus. Extensions and 
applications to areas such as information retrieval are 
suggested. 
1 Introduction 
Currently there is much interest in the automatic ac- 
quisition of lexiea\[ syntax and semantics, with the 
goal of building up large lexicons for natural lain 
guage processing. Projects that center around ex- 
tracting lexical information from Machine Readable 
Dictionaries (MRDs) have shown much success but 
are inherently limited, since the set of entries within 
a dictionary is fixed. In order to find terms and ex- 
pressions that are not defined in MRDs we must turn 
to other textual resources. For this purpose, we view 
a text corpus not only as a source of information, but 
also as a source of information about the language it 
is written in. 
When interpreting unrestricted, domain-independent 
text, it is difficult to determine in advance what kind 
of infbrmation will be encountered and how it will be 
expressed. Instead of interpreting everything in the 
text in great detail, we can searcil for specific lexical 
relations that are expressed in well-known ways. Sur- 
prisingly useful information can be found with only 
a very simple understanding of a text. Consider the 
following sentence: 1. 
(SI) The bow lute, such as the Bambara ndang, 
is plucked and has an individual 
curved neck :for each string. 
Most fluent readers of English who }lave never be- 
fore encountered the term 'q3amhara ndang" will nev- 
ertheless from this sentence infer that a "Bambara 
udang" is a kind of "bow Iute". This is true even if 
tile reader has only a fuzzy conception of what a how 
lute is. Note that the attthor of the sentence is not de- 
liberately defining the term, as would a dictionary or 
a children's book containing a didactic sentence like 
A Bambara ndang is a kind of bow lute. However, 
the semantics of the lexico-syntactic construction in- 
dicated by the pattern: 
(la) NPo ..... h as {NP1, NP2 .... (and Ior)} NP,, 
are such that they imply 
(lb) for all NP,, 1 < i< n, hyponym(NPi, NPo) 
Thus from sentence (SI) we conclude 
hyponym ( "Barn bare n dang", "how lu re"). 
We use the term hyponym similarly to the sense used 
in (Miller et el. 1990): a concept represented by a 
lexicaI item L0 is said to be a hyponym of the concept 
represented by a lexical item LI if native speakers of 
English accept sentences constructed from the frame 
An Lo is a (kind of) L1. Here Lt is the hypernym 
of Lo and the relationship is reflexive and transitive, 
but not symmetric. 
This example shows a way to discover a hyponymic 
lexical relationship between two or more noun phrases 
in a naturally-occurring text. This approach is siml- 
lar in spirit to the pattern-based interpretation tech- 
niques being used in MRD processing. For example, 
t All examples in this paper are real text, taken from 
Grolter's Amerwan Acaderntc Encyclopedia(Groher tg00) 
AcrF.s DE COLING-92, NANTI~S, 23-28 Aol}r 1992 5 3 9 PROC. OV COLING-92, NhNTIIS, AUG. 23-28, 1992 
(Alshawi 1987), in interpreting LDOCE definitions, 
uses a hierarchy of patterns which consist mainly 
of part-of-speech indicators and wildcard characters. 
(Markowitz e~ al. 1986), (Jensen & Binot 1987), and 
(Nakamura & Nagao 1988) also use pattern recogni- 
tion to extract semantic relations such as taxonomy 
from various dictionaries. (Ahlswede & Evens I988) 
compares an approach based on parsing Webster's 
7th definitions with one based on pattern recognition, 
and finds that for finding simple semantic relations, 
pattern recognition \[s far more accurate and efficient 
than parsing. The general feeling is that the struc- 
ture and function of MRDs makes their interpretation 
amenable to pattern-recognition techniques. 
Thus one could say by interpreting sentence (S1) ac- 
cording to (In-b) we are applying pattern-based rela- 
tion recognition to general texts. Since one of the 
goals of building a lexical hierarchy automatically 
is to aid in the construction of a natural language 
processing program, this approach to acquisition is 
preferable to one that needs a complex parser ~nd 
knowledge base. The tradeoff is that the the refor- 
mation acquired is coarse-grained. 
There are many ways that the structure of a lan- 
guage can indicate the meanings of lexical items, but 
the difficulty lies in finding constructions that fre- 
quently and reliably indicate the relation of interest. 
It might seem tbat because free text is so varied in 
form and content (as compared with the somewhat 
regular structure of the dictionary) that it may not 
be possible to find such constructions. However, we 
have identified a set of lexico-syntactic patterns, in- 
cluding the one shown in (In) above, that indicate 
the hyponymy relation and that satisfy the following 
desiderata: 
(i) They occur frequently and in many text genres. 
(ii) They (almost) always indicate the relation of in- 
terest. 
(iii) They can be recognized with little or no pre- 
encoded knowledge. 
Item (i) indicates that the pattern will result in the 
discovery of many instances of the relation, item (ii) 
that the information extracted will not be erroneous, 
and item (iii) that making use of the pattern does not 
require the tools that it is intended to help build. 
Finding instances of the hyponymy relation is useful 
for several purposes: 
Lexicon Augmentation. Hyponymy relations can 
be used to augment and verify existing lexicons, in- 
cluding ones built from MRDs. Section 3 of this 
paper describes an example, comparing results ex- 
tracted from a text corpus with information stored in 
the noun hierarchy of WordNet ((Miller et al. 1990)), 
a hand-built lexical thesaurus. 
Noun Phrase Semantics. Another purpose to 
which these relations can be applied is the identifi- 
cation of the general meaning of an unfamiliar noun 
phrases. For example, discovering the predicate 
hyponym( "broken bone", "injury") 
indicates that tbe term "broken bone" can be under- 
stood at some level as an "injury" without having to 
determine the correct senses of the component words 
and how they combine. Note also that a term like 
"broken bone" is not likely to appear in a dictionary 
or lexicon, although it is a common locution. 
Semantic Relatedness Information. There bas 
recently been work in the detection of semantically re- 
lated nouns via, for example, shared argument struc- 
tures (Hindle 1990), and shared dictionary definition 
context (Wilks e¢ al. 1990). These approaches at- 
tempt to infer relationships among \[exical terms by 
looking at very large text samples and determining 
which ones are related in a statistically significant 
way. The technique introduced in this paper can be 
seen as having a similar goal but an entirely different 
approach, since only one sample need be found in or- 
der to determine a salient relationship (and that sam- 
ple may be infrequently occurring or nonexistent). 
Thinking of the relations discovered as closely related 
semantically instead of as hyponymic is most felic- 
itous when the noun phrases involved are modified 
and atypical. Consider, for example, the predicate 
hyponym( "detonating explosive", "blasting agent") 
This relation may not be a canonical ISA relation but 
the fact that it was found in a text implies that the 
terms' meanings are close. Connecting terms whose 
expressions are quite disparate but whose meanings 
are similar should be useful for improved synonym ex- 
pansion in information retrieval and for finding chains 
of semantically related phrases, as used in the ap- 
proach to recognition of topic boundaries of (Morris 
Hirst 1991). We observe that terms that occur in a 
list are often related semantically, whether they occur 
in a hyponymy relation or not. 
In the next section we outline a way to discover these 
lexico-syntactic patterns as well as illustrate those we 
have found. Section 3 shows the results of searching 
texts for a restricted version of one of the patterns and 
compares the results against a hand-built thesaurus. 
Section 4 is a discussion of the merits of this work 
and describes future directions. 
2 Lexico-Syntactic Patterns 
for Hyponymy 
Since only a subset of the possible instances of the 
hyponymy relation will appear in a particular form, 
we need to make use of as many patterns as possi- 
ble. Below is a list of lexico-syntactie patterns that 
indicate the hyponymy relation, followed by illustra- 
tive sentence fragments and the predicates that can 
ACTI~S DE COLING-92, NANTES, 23-28 AOt~r 1992 5 4 0 PROC. OF COLING-92, NANTES, AUG. 23-28, 1992 
be derived from them (detail about the environment 
surrounding tile patterns is omitted for simplicity): 
(2) .... h NP us {NP ,}* {(or \[ and)} NP 
... works by such authors as Herrick, 
Goldsmith, and Shakespeare. 
: ~. hyf)onym I'~author", "Ilerrick'), 
llyponym( "author", "(;oldsmith "), 
hyponynl( "author", "Shakespeare") 
(3) NP {, NP} * {,} o,' other NP 
Bruises, wounds, broken bones or other 
injuries . . . 
~... hyponym( "bruise". "injury"), 
hyponym ( "wo und", "mj ury" ), 
hyponym( "broken bone", "injury") 
(4) NP {, NP}* {,} and other NP 
... temples, treasuries,altd other 
important civic buildings. 
:~- hyponym("tenlple", "civic' building"), 
hyponym( "treasury ", "civic building") 
(5) m, {,} .~clsa,,~y {NP 5* {o,. ' ..a} NP 
All common-law countries, including 
Canada and England ... 
-~, hyponym( "Canada", "collllnou--law coon 
try"), flyponym ( "Eng\]and", "common-law co lm - 
try") 
(6) NP {,} especially {NP ,}* {or\] and} NP 
... most: European countries, especially 
France, England, and Spain. 
~ hyponym( "France", "European country"), 
hyponym( "England", "European country"), 
hypouym( "Spain", "European country") 
When a relation hyponym(NPo, NI'I) is discov- 
ered, aside from some temmatizing and removal of 
unwanted modifiers, tile uonn phrase is left as all 
atomic unit, not broken clown and analyzed. Ira more 
detailed interpretation is desired, the results can be 
passed on to a more intelligent or specialized language 
analysis component. And, as mentioned above, this 
kind of discovery procedure can be a partial solution 
for a problenr like noun phrase interpretation because 
at least part of the meaning of the phrase is indicated 
by tile hyponymy relation. 
and we usually want them to be singular. Adjecti- 
val quantiflers such as "other" and "some" are usu- 
ally undesirable and can be eliminated in most cases 
without making the statement of tile hypouym rela- 
tion erroneous. ('omparatives SUCh as "inlportaat" 
and "smaller" are usually best removed, since their 
meaning \[s relative and dependent on tile context in 
which they appear. 
Ilow much modification is desirable depends on the 
application to which the lexical relations will be put. 
For budding up a basic, general-domain thesaurus, 
single-word uouns and very cOnllnon colnpouuds are 
most appropriate. For a inore specialized domain, 
umre modified terms have their place. Per example, 
noun phrases in ~he me(licai ¢lontain otteu have sev- 
eral layers of modification which should be preserved 
in a taxonomy of medical terms. 
Other difficulties and concerns are discussed ill Sec- 
tion a. 
2.2 Discovery of New Patterns 
How can these patterns be found? Initially we dis- 
covered patterns (1)- (3) 5y observation, looldug 
through text and noticing die patterns and tile rela- 
tionships they indicate, lu order to find new patterns 
automatically, we sketch the following procedure: 
1. l)ecide on a lexical relation, R, that is of interest, 
e.g., "gro up/member"(iu our formulation this is 
a subset of the hypouylny relation). 
2. Gather a list of terms for which this rela- 
tion is known to hold, e.g., "England-country'. 
This list can be found autonmtically using the 
method described here, bootstrapping from pat- 
terns found by hand, or by bootstrapping from 
an existing lexicon or knowledge base. 
3. Find places in tile corpus where these expressions 
occur syntactically near one another and record 
the environment. 
4. t,'ind the commonaflties among these environ- 
i~leuts and hypothesize that corn.men ones yield 
patterns that indicate the relation of interest. 
5. Once a new pattern has been positively identi- 
fied, use it to gather more instances of the target 
relation and go to Step 2. 
2.1 Some Considerations 
In example (4) above, the full noun phrase corre- 
sponding to the hypernym is "other important civic 
buildings". This illustrates a difficulty that arises 
from using free text as the data source, as opposed 
to a dictionary - often the form that a noun phrase 
occurs in is not what we would like to record. For 
example, nouns frequently occur in their plural form 
We tried this procedure by hand using just one pair 
of terms at a time. In the first case we tried the 
"Fngland-country" example, and with just this pair 
we tound uew patterns (4) and (5), as well as (1) 
(3) which were already known. Next we tried "tank- 
vehicle" and discovered a very productive pattern, 
pattern (6). (Note that for this pattern, even though 
it has an emphatic element, this does not affect the 
fact that the relation indicated is hypouymic.) 
AcrEs DE COLING-92, N^mEs, 23-28 hotrr 1992 5 4 1 l)Roc, ov COLING-92, NAbrrEs, AUG. 23-28, 1992 
We have tried applying this technique to meronymy 
(i.e., the part/whole relation), but without great suc- 
cess. The patterns fotu~.d for this relation do not tend 
to uniquely identify it, but can be used to express 
other relations as well. It may be the case that in 
English the hyponymy relation is especially amenable 
to this kind of analysis, perhaps due to its "naming" 
nature. However, we have bad some success at iden- 
tification of more specific relations, such as patterns 
that indicate certain types of proper nouns. 
We have not implemented an automatic version of 
this algorithm, primarily because Step 4 is underde- 
termined. 
2.3 Related Work 
This section discusses work in acquisition of lexical in- 
formation from text corpora, although as mentioned 
earlier, significant work has been done in acquiring 
lexical information from MRDs. 
(Coates-Stephens 1991) acquires semantic descrip- 
tions of proper nouns in a system called FUNES. FU- 
NES attempts to fill in frame roles, (e.g., name, age~ 
origin, position, and works-for, for a person frame) 
by processing newswire text. This system is simi- 
lar to the work described here in that it recognizes 
some features of the context in which the proper noun 
occurs in order to identify some relevant semantic 
attributes. For instance. Coates-Stephens mentions 
that "known as" can explicitly introduce meanings 
for terms, as can appositives. We also have consid- 
ered these markers, hut the tbrmer often does not 
cleanly indicate "another name for" and the latter is 
difficult to recognize accurately. FUNES differs quite 
strongly from our approach in that, because it is able 
to fill in many kinds of frame roles, it requires a parser 
that produces a detailed structure, and it requires a 
domain-dependent knowlege base/lexicon. 
(Velardi & Pazienza 1989) makes use of hand-coded 
selection restriction and conceptual relation rules in 
order to assign case roles to lexical items, and (Ja- 
cobs & Zernik 1988) uses extensive domain knowledge 
to fill in missing category information for unknown 
words. 
Work on acquisition of syntactic information from 
text corpora includes Brent's (Brent 1991) verb 
subcategorization frame recognition technique and 
Smadja's (Smadja & McKeown 1990) collocation ac- 
quisition algorithm. (Calzolari & Bindi 1990) use 
corpus-based statistical association ratios to deter- 
mine lexical information such as prepositional com- 
plementation relations, modification relations, and 
significant compounds. 
Our methodology is similar to Brent's in its effort 
to distinguish clear pieces of evidence from ambigu- 
ous ones. The assumption is that that given a large 
enough corpus, the algorithm can afford wait until 
it encounters clear examples. Brent's algorithm re- 
lies on a clever trick: in the configuration of interest 
(in this case, verb valence descriptions), where noun 
phrases are the source of ambiguity, it uses only sen- 
tences which have pronouns in the crucial position, 
since pronouns do not allow this ambiguity. This 
approach is qnite effective, but the disadvantage is 
that it isn't clear that it is applicable to any other 
tasks. The approach presented in this paper, using 
the algorithm sketched in the previous subsection, is 
potentially extensible. 
3 Incorporating Results into 
WordNet 
To validate this acquisition method, we compared the 
results of a restricted version of the algorithm with 
information found in WordNet. 2 WordNet (Miller 
et al. 1990) is a hand-built online thesaurus whose 
organization is modeled after the results of psycbolin- 
guistic research. To use tile authors' words, Wordnet 
"... is an attempt to organize lexical information in 
terms of word meanings, rather than word forms. In 
that respect, WordNet resembles a thesaurus more 
than a dictionary ..." To this end, word forms with 
synonymous meanings are grouped into sets, called 
synsets. This allows a distinction to be made be- 
tween senses of homographs. For example, the noun 
"board" appears in the synsets {board, plank} and 
{board, committee}, and this grouping serves for the 
most part as the word's definition. In version 1.1, 
WordNet contains about 34,000 noun word forms, 
including some compounds and proper nouns, orga- 
nized into about 26,000 synsets. Noun synsets are 
organized hierarchically according to the hyponymy 
relation with implied inheritance and are further dis- 
tinguished by values of features such as meronymy. 
WordNet's coverage and structure are impressive and 
provide a good basis for an automatic acquisition al- 
gorithm to build on. 
When comparing a result hyponym(No,Nt) to the 
contents of WordNet's noun hierarchy, three kinds of 
outcomes are possible: 
Verify. If both No and Nt are in WordNet, and if the 
relation byponym(No,N1) is in the hierarchy (possi- 
bly througi~ transitive closure) then the thesaurus is 
verified. 
Critique. If both No and N1 are in WordNet, and if 
the relation hyponym(No, N1) is not in the hierarchy 
(even through transitive closure) then the thesaurus 
is critiqued, i.e., a new set of hyponym connections is 
suggested. 
Augment. If one or both of No and NI are not 
present then these noun phrases and their relation 
are suggested as entries. 
As an example of critiquing, consider the following 
2The author thanks Miller, et al,, for the distribution of 
WordNet. 
AcrEs DE COL1NG-92, NANTES, 23-28 AoU'r 1992 5 4 2 PRec. OF COLING-92, NANTES, AUG. 23-28, 1992 
sentence and derived relation: 
(S2) Other input-output dev±ces, such as 
printers, color plotzers, ... 
~ hyponym('~rinter','~npnt-mltput device") 
The text indicates that a printer is a kind of input- 
output device. Figure 1 indicates tile portion of tile 
hyponymy relation in WordNet's noun hierarchy that 
has to do with printers and devices. Note ;although 
the terms device and printer are present, they are not 
linked in such as way as to allow the easy insertion 
UO device under the more general dewce and over the 
more specific printer. Although it is not obvious what 
to suggest to fix this portion of the hierarchy from 
this one relation ~done, it is clear that its discovery 
highlights a trouble spot ill tile structure. 
,__/_"-._._, 
Figure t: A Fragment of the WordNet Noun Hier- 
archy. Syasets are enclosed in braces; most synsets 
have more connections than those shown. 
aereal~: ricu* ~heat* 
countries: Cuba Vietnam France* 
hydrocarbon: ethylene 
~ubstances: bromine* hydrogen* 
protozoa: parameclum 
liqueurs: anisette* absinthe* 
rocks: graltlte* 
substances: phosphorus* nitrogen* 
species: stuatornis oilbirds 
bivalves: scallop* 
fungi: smuts* rusts* 
fabrics: acrylics* nylon* silk* 
antibiotlcS: amplcillin erythromycln* 
institutions: temples king 
seabirds: penguins albatross* 
flatworms: tapeworms pla~aria 
amphibians: frogs* 
~aterfowl: ducks 
legumes: lentils* beans* nuts 
org~lisms: horsetails ferns mosses 
rivers: Sevier Ca\[rson Humboldt 
fruit: olives* grapes* 
hydrocarbons: benzene gasol±ne 
ideologies: liberalism conservatism 
industries: steel iron shoes 
min.rals: pyrite* galena 
phenomena: lightning* 
infection; menlngltis 
dyes: quercitron 
Figure 2: Relations found in Grolier's. The format 
is hypernym: hyponyrn list. Entries with * indicate 
relations found in WordNet. 
Most of the terms in WordNet's noun hierarchy are 
unmodified nouns or nouns with a single modifier. 
For this reason, ill this experiment we only extracted 
relations consisting of mmmdified nouns in both the 
hypernym and hypouym roles (although determiners 
are allowed and a very small set of quantifier ad- 
jectives: "some", "many", "certain", and "other"). 
Making this restriction is also usethl because of the 
difficulties with determining which modifiers are sig- 
nificant, as touched on above, and because it seems 
easier to make a judgement call about the correctness 
of the classification of unmodified nouns for evalua- 
tion purposes. 
Since we are trying to acquire lexical information our 
parsing mechanism should not be one that requires 
extensive lexicat information. In order to detect the 
lexico-syntactic patterns, we use a unification-based 
constituent analyzer (taken from (Batali 1991)), 
which builds on the output of a part-or=speech tag- 
ger (Cutting el al. 1991). (All code described in this 
report is written m Common Lisp and run on Sun 
SparcStations.) 
We wrote grammar rules for the constituent analyzer 
to recognize the pattern in (la). As mentioned above, 
in this experiment we are detecting only unmodified 
nouns. Therefore, when a noun is found in the hyper- 
nym position, that is, before the lexemes "such as", 
we check for the noun's inclusion in a relative clause, 
or as part of a larger noun phrase that includes an 
appositive or a parenthetical. Using tile constituent 
analyzer, it is not necessary to parse the entire sell- 
tence; instead we look at just enough local context 
around the iexical items in the pattern to ensure that 
tile nouns in tile pattern are isolated. 
After the hypernym is detected the hyponyms are 
identified. Often they occur ill a llst and each ele- 
ment ill the list holds a hyponym relation with the 
hypernym. The main difficulty here lies m determin- 
ing the extent of the last term in the list. 
3.1 Results and Evaluation 
Figure 2 illustrates some of the results of a run of 
the acquisition algorithm on Grolier's American Aca- 
demic Encyelopedia(Grolier 1990), where a restricted 
version of pattern (la) is the target (space constraints 
do not allow a full listing of the results). After the re- 
lations are found they are looked up in WordNet. We 
placed the WordNet noun hierarchy into a b-tree data 
structure for efficient retrieval and update and used a 
breadth-first-search to search through the transitive 
closure. 
Ont of 8.6M words of encyclopedia text, there are 
AcrEs DE COL1NG-92, NANt .'F.S, 23-28 ho,,~'r 1992 5 4 3 Paoc. ov COLING-92, NANTES, AUO. 23-28, 1992 
7067 sentences that contain tile lexemes "such as" 
contiguously. Out of these, 152 relations fit tile re- 
strictions of the experiment, namely that both the 
hyponyms and the hypernyms are unmodified (with 
the exceptions mentioned above). When the restric- 
tions were eased slightly, so that NPs consisting of 
two nouns or a present/past participle plus a noun 
were allowed, 330 relations were found. Wheu the lat- 
ter experiment was run o21 about 20M words of New 
York Times text, 3178 sentences contained "such as" 
contiguously, and 46 relations were found using the 
strict no-modifiers criterion. 
Wilen the set of t52 Grolier's relations was looked up 
in WordNet, 180 out of the 226 mlique words involved 
in the relations actually existed in the hierarchy, and 
61 out of the 106 feasible relations (i.e., relations in 
which both terms were already registered in Word- 
Net) were found. 
The quality of the relations found seems high over- 
all, although there are difficulties. As to be expected, 
metonymy occurs, as seen in hyponym("king", "in- 
stitution"). A more common problem is under- 
specification. For example, one relation is hy- 
ponym( "steatornis', "species"), which is problematic 
because what kind of species needs to be known and 
most likely this reformation was mentioned in the pre- 
vious sentence. Similarly, relations were found be- 
tween "device" and "plot", "metaphor", and "char- 
acter", underspecifying the fact that literary devices 
of some sort are under discussion. 
Sometimes the relationship expressed is slightly 
askance of the norm. For example, the algorithm 
finds hyponym( "Washington", "nationalist")and hy- 
ponym( "aircraft", "target") which are somewhat con- 
text and point-of-view dependent. This is not neces- 
sarily a problem; as mentioned above, finding alter- 
native ways of stating similar notions is one of our 
goals. However, it is important to try to distinguish 
the more canonical and context-independent relations 
for entry in a thesaurus. 
There are a few relations whose hypernyms are very 
high-level terms, e.g., "substance" aud "form". These 
are not incorrect; they just may not be as useful as 
more specific relations. 
Overall, the results are encouraging. Although the 
number of relations found is small compared to the 
size of the text used, this situation can he greatly im- 
proved in several ways. Less stringent restrictions will 
increase the numbers, as the slight loosening shown 
in the Grolier's experiment indicates. A more savvy 
grammar for the constituent analyzer should also in- 
crease the results. 
3.2 Automatic Updating 
The question arises as to how to automatically in- 
sert relations between terms into the hierarchy. This 
involves two main difficulties. First, if both lexical 
expressions are present in the noun hierarchy but one 
or both }lave more than one sense, the algorithm must 
decide which senses to link together. We have prelim- 
inary ideas as to how to work around this problem. 
Say the hyponym in question has only one sense, but 
the hypernym has several. Then the task is simplified 
to determining which sense of the hypernym to link 
the hypouym to. We can then make use of a lexical 
disambiguation algorithm, e.g., (Hearst 1991), to de- 
termine which sense of the hypernym is being used iu 
the sample sentence. 
Furthermore, since we've assumed the hyponym has 
only one main sense we could do tile following: Look 
through a corpus for occurrences of the hyponym and 
see if its environment tends to be similar to one of the 
senses of its hypernym. For example, if the hypernym 
is "bank" and the hyponym is "First National", ev- 
ery time, within a sample of text, the term "First 
National" occurs, replace it with "bank", and then 
run the disambiguation algorithm as usual. If this 
term can be positively classified as having one sense of 
bank over the others, then this would provide strong 
evidence as to which sense of the hypernym to link 
the hypouym to. This idea is purely speculative; we 
have not yet tested it. 
The second main problem with inserting new rela- 
tions arises when one or both terms do not occur in 
the hierarchy at all. In this case, we have to deter- 
mine which, if any, existing synset the term belongs 
in and then do the sense determination mentioned 
above. 
4 Conclusions 
We have described a low-cost approach for automatic 
acquisition of semantic lexical relations from uure- 
stricted text. This method is meant to provide an 
incremental step toward the larger goals of natural 
language processing. Our approach is complementary 
to statistically based approaches that find semantic 
relations between terms, iu that ours requires a sin- 
gle specially expressed instance of a relation while 
the others require a statistically significant number 
of generally expressed relations. We've shown that 
our approach is also useful as a critiquing component 
for existing knowledge bases and lexicons. 
We plan to test the pattern discovery algorithm on 
more relations and on languages other than English 
(depending on the corpora available). We would also 
like to do some analysis of the noun phrases that are 
acquired, and to explore the effects of various kinds of 
modifiers on the appropriateness of the noun phrase. 
We plan to do this in the context of analyzing envi- 
ronmental impact reports. 
Acknowledgements. This work was supported in 
part by an internship at tile Xerox Palo Alto Research 
Center and in part by the University of California and 
Digital Equipment Corporation under Digital's flag- 
AcrEs DE COLING-92, NANTES, 23-28 ^o~-r 1992 5 4 4 PRoc. OF COLING-92. NANTES, Auo. 23-28, 1992 
ship research project Sequoia 2000: Large Capacity 
Object Servers to Support Global Change Research. 

References 

Ahlswede, T. & M. Evens (1988). Parsing vs. text 
processing in the analysis of dictionary defini- 
tions. Proceedings of the 26th Annual Meeting of 
the Association for Computational Linguistics, 
pages 217-224. 

Alshawi, H. (1987). Processing dictionary definitions 
with phrasal pattern hierarchies. American Jour- 
nal of Computational Linguistics, 13(3):195 202. 

Batali, J. (1991). Automatic Acquisition and Use of 
Some of the Knowledge in Physics Tezts. PhD 
thesis, Massachusetts Institute of Technology, 
Artificial Intelligence Laboratory. 

Brent, M. R. (1991). Automatic acquisition ofsubcat- 
egorization frames from untagged, free-text cor- 
pora. In Proceedings of the 29th Annual Meet- 
ing of the Association fo'e Computational Lin- 
guistics. 

Calzolari, N. & R. Bindi (1990). Acquisition of lexi- 
cal information from a large textual italian cor- 
pus. In Proceedings of the Thirteenth Interna- 
tional Conference on Computational Linguistics, 
Helsinki. 

Coates-Stephens, S. (1991). Coping with lexical in- 
adequacy - the automatic acquisition of proper 
nouns from news text. In The Proceedings of the 
7th Annual Conference of the UW Centre for the 
New OED and Tezt Research: Using Corpora, 
pages 154-169, Oxford. 

Cutting, D., J. Kupiec, J. Pedersen, & P. Sibun 
(1991). A practical part-of-speech tagger. Sub- 
mitted to The 3rd Conference on Applied Natural 
Language Process*ng. 

Grolier (1990). Academic American Encyclopedia. 
Grolier Electronic Publishing, Danbury, Con- 
neeticut. 

Jensen, K. & J.-L. Binot (1987). Disambiguating 
prepositional phrase attachments by using on- 
line dictionary definitions. American Journal of 
Computational Linguistics, 13(3):251-260. 

Markowitz, J., T. Ahlswede, & M. Evens (1986). Se- 
mantically significant patterns in dictionary def- 
initions. Proceedings of the 24th Annual Meet- 
ing of the Assoczation for Computational Lin- 
guistics, pages 112-119. 

Miller, G. A., R. Beckwith, C. Fellbaum, D. Gross, & 
K. J. Miller (1990). Introduction to wordnet: An 
on-line lexieal database. Journal of Le~xieography, 
3(4):235-244. 

Morris, J. & G. Hirst (1991). Lexical cohesion com- 
puted by tbesaural relations as an indicator of 
the structure of text. Computational Lzngmstics, 
17(1):21-48. 

Nakamura, J. & M. Nagao (1988). Extraction of se- 
mantic inlbrmation t¥om an ordinary english dic- 
tionary and its evaluation. In Proceedings of the 
Twelfth International Conference on Computa- 
tional Linguistics, pages 459-464, Budapest. 

Smadja, F. A. & K. R. McKeown (1990). Automati- 
cally extracting and representing collocations for 
language generation. Proceedings of the 28th An- 
nual Meeting of the Association for Computa- 
tional Linguistics, pages 252-259. 

Velardi, P. & M. T. Pazienza (1989). Computer aided 
interpretation of lexical cooccurrences. Proceed- 
ings of the 27th Annual Meeting of the Associ- 
ation for Computational Linguistics, pages 185-192. 

Wilks, Y. A., D. C. Fass, C. ruing Guo, J. E. McDon- 
ald, T. Plate, & B. M. Slator (1990). Providing 
machine tractable dictionary tools. Journal of 
Machzne Translation, 2. 

Hearst, M. A. (1991). Noun homograph disambigua- 
tion using local context in large text corpora. In 
The Proceedings of the 7th Annual Conference 
of the UW Centre for the New OED and Tezt 
Research: Using Corpora, Oxford. 

Hindle, D. (1990). Noun classification from predicate- 
argument structures. Proceedings of the 28th An- 
nual Meeting of the Association for Computa- 
tional Linguistics, pages 268-275. 

Jacobs, P. & U. Zernik (1988). Acquiring lexical 
knowledge from text: A case study. In Proceed- 
ings of AAAI88, pages 739-744. 
