EBL2: AN APPROACH TO 
AUTOMATIC LEXICAL ACQUISITION 
LARS ASKER* BJ()R N GAM BACK ~ CllRISTER SAM UELSSON 1 
asker@day, nu. ne gam©sics, so christ er~sics, se 
Keywords: linguistic tools: /exical acquisiti(ul; explanation-based learning 
Abstract 
A method for automatic lexical acquisition is out 
lined. An existing lexicon that, in addition Io ordi- 
nary \]exical entries, contains prototypical cntrips for 
various non-exclusive paradigms of open-cl~,.ss words, 
is extended by inferring new lexical entries from texts 
containing unknown words. This is done by com- 
paring the constraints placed on the unknown words 
hy the natural language system's grammar with the 
prototypes and a number of hand-coded phras(, tem- 
plates specific for each paradigm. Once a sufficient 
number of observations of the word in different con- 
texts have been made, a lexical entry is constructed 
for the word by assigning it to one or sew~ral para- 
digm(s), 
Parsing sentences with ullknown words is nor- 
mally very time-consuming due to the large nmn- 
ber of grammatically possible analyses. To cir~ 
cumvent this problem, other ilhrase templates are 
extracted automatically from tim gramlnal and 
domain-specific texts using an explanation based 
learning method. These templates represent gram- 
matically correct, sentence patterns. When a sell- 
tence matches a template, the original I)arsing com- 
ponent can be bypassed, reducing parsing times dra- 
matically. 
1 Introduction 
A persisting trend in unification-based approaches to 
natural language processing is to incorporate large 
quantirAes of information in the lexicon, informatio,i 
that has traditionally resided in the gran,mar rules. 
Acquiring a lexicon has thus becolne a diflicull and 
time consuming txsk, even for moderately sized lex 
ira. In addition to this, an.',' natural language pro- 
cessing system intended for serious applications must 
include a large lexicon -- several thousands of words 
or more is commonly considered a minimun~ size 
which adds even more to the complexity of the lu'ob- 
lem. In view Of this, tools for lexical acqusition are 
not only desirable they become a necessity. 
Most. approaches to this problem hay,' been 
*Department of (.\]onlpuler ,~lld S)'steRIS ScieIIC('S, S\[ock. 
hohn University, Electrnm 23(/, S - 16.t ,In \]<.ISTA, Sweden. 
I NLP-group, Swedish hmtitute tff Computer Science, Box 
1263, S 16.1 28 KIST& Stockhohn, Sweden. 
to construct a range of tools that require vari- 
nus degrees of inleraclive support when new lexi- 
cal entries are created, either from raw text ma- 
terial (as ill e.g., \['frost & Bnchberger 86, Grosz ct 
al 87, Wilensky 90\] and tile early work by Zernik 
\[Zernik ~(," Dyer 85, Zernik 87\]), or from machine 
readable dictionaries (see e.g., \[Bognraev el al 87, 
('.alzolari &" Bindi 90\]). Although interactive tools 
|or h'xical acquisition greatly simplifies tile task of 
constructing a lexicon, it. is desirable to go oue step 
further and fully remow" the need for user interac- 
tion. 
One of the first systems that aimed at construct- 
ing lexica\] entries automatically from raw text was 
Granger's FOUL-U'P system \[Granger 77\]. FOUL- 
UP extended a lexicon by referring restrictions 
placed on unknown words by instantiating scripts 
that matched the sentences containing the nnknown 
words. This I)uilt on a immber of assumptions which 
in general do nol bold, in particular: that all the 
information needed to create all entry is contained 
ill one text: that no nmrphological information is 
needed; tha~ specific (hand-coded) scripts covering 
the domain can be made available in advance, hi 
one of the later approaches to automatic lexical ac- 
quisition from raw text, \[dacobs ,to Zernik 88\] have 
shown the need to consult a variety of knowledge 
sources such ~s morphological, syntactic, semantic, 
and contextual knowledge when determining a new 
lexical entry. 
This paper describes an automatic nlethod to ac- 
quire new lexical entries by using analytical learning 
in coml,inalion wit.h strategies used in an existing 
interactive tool for lexical acquisition (VEX \[Carter 
89}). In the process of constructing a lexical en- 
try. the system combines several different sources of 
information: the underlying NL system (CLE, \[Al- 
shawi red.) 92\]) will contribute information on syn- 
tactically and semantically permissible phrases and 
on tile rules for inIlectional nmrl)hology. The corpus 
wilt contrihute information on which of these con- 
structions actually occur. This information is com- 
bined with tile the linguistic knowledge encoded in 
the interactive lexical acquisition tool to infer lexical 
entries for unknown words m the text. 
The rest of Ihe paller is laid out as follows: Sec- 
tion :2 contains information al)out the various ele- 
ments on which the method is based. Section 3 de- 
AcrEs DE COLING-92, NANTES, 23-28 Aot)'r 1992 1 l 7 2 PRec. el: COLING-92, NANTES, AUG. 23-28, 1992 
scribes the method itself and Section 4 reports on For these "paradigm words" only, the complete set 
the current state of the implementation, of feature vahles is explicitly specified. 
2 The elements of the scheme 
2.1 The Core Language Engine, CLE 
The Core Language Engine is a general purpose nat- 
ural language processing system for English devel- 
oped by SRI Cambridge. It is intended to be used 
as a building block in a broad range of applications, 
e.g. data-b~.se query systems, machine translation 
systems, text-to-speecb/speech-to-text systems, etc. 
The object of the CLE is to map certain natural 
language expressions into appropriate predicates in 
logical form (or Quasi-Logical Form \[Alshawi ,(.: van 
Eijck 89\]). The system is based completely on tmi 
lication and facilitates a reversible phrase-structure 
type grammar. 
The Swedish Institute of Computer ,'qci(m(e has 
with support from 8RI generalized the fi'anwwork 
and developed all equivahmt system for Swedish (the 
S-CLE, \[Gamback & Rayner 92\]). The two copies of 
the CLE have been used together to form a machine 
translation system \[Alshawi et a191\]. The S-('LE has 
a fairly large gramnmr covering most of the common 
constructions in Swedish. There is a good treatment 
of inflectional morphology, covering all main inflec- 
tional closes of nouns, verbs and adjectives. 
The wide range of l)ossihle applications have put 
severe restrictions on the type of lexicon that can 
be used. The S-CLE h~ a function-word lexico~J 
containing about 400 words, including most Swedish 
pronouns, conjllnctlous, prepositions, determiners, 
particles and "special" verbs. In addition, there is 
a "core" content-word lexicon (with common nouns, 
verbs and adjectives) and domain specitic h'xica. 
This part of tbe system is still under development 
and all these content-word lexica together haw, about 
750 entries. 
The lexical entries contain information about il~- 
flectional morphology, syntactic and semantic sub- 
categorization, anti sortal (selectional) restrictions. 
Information abont the linguistic properties of an en- 
try is represented by complex categories that include 
a principal category symbol and specifications of con- 
straints on the values of syntactic/semantic features. 
Such categories also appear in the C.LF,'s grammar 
and matching and merging of the information en- 
coded in them is carried out by unification during 
parsing. Two categories can be unified if the con- 
straints on their feature values are compatible 
In the actual "core" and domain Icxica, this infor- 
mation is kept implicit and represented as pointers 
to entries in a "paradigm" lexicon with a number of 
words representing basic word usages and inflections. 
2.2 The Vocabulary EXpander, VEX 
In the English CLE, new lexicon entries can be added 
by tile users with a tool developed for the purpose. 
q'his lexicon acquisition tool, the Vocabulary EX- 
pander, is fully described in \[Carter 89\]. In parallel 
with the development of the S-CLE, a Swedish ver- 
sion of the VEX system was designed \[Gamback 92\]. 
VEX allows for the creation of lexical entries by 
users with knowledge both of a natural language and 
of a Sl)ecilic application domain, but not of linguistic 
theory or of tile way lexical entries are represented in 
the CLE. It presents examl)le sentences to the user 
and asks lor information on tile grammaticality of 
the sentences, and for selcctional restrictions on ar- 
guments of predicates VEX adopts a copy and edit 
strategy in colmtrnctiug Icxical entries. It builds on 
the "paradigm" lexicon and sentence patterns, that 
is, declarative knowledge of the range of sentential 
contexts ill which the word usages in that lexicon 
Call OCCUI'. 
In the present work we want to investigate to 
what extent snch creation of lexicon entries can be 
performed with a minimum of user interaction, ln- 
stead of presenting exaruple sentences to the user we 
are allowing the program to use a very large text 
where hopefully unknown words will occur in sev- 
eral ditlbrenl sentence patterns. This strategy will 
he filrther described i~, the following sections. 
First, however, we will define what we mean by 
the notion of (subcategorization) "paradigm". Tile 
definition we adopt here is based on the one used in 
\[Carter 89\], namely that 
Definition 1 
a paradigm zs any minimal non.empty intersection 
of Icxical entries. Every category in a pa,'adlgm will 
occur in czaclly the same set of entries in the lexicon 
as every other category Of auy) in that paradigm. 
Every ent,y consists of a dis3o2ul union of paradigms. 
lh're, we assume that a lexicon can be described 
in terms of (a small set of) sucb paradigms, relying 
on ttle fact. that the open-class words exhibit at least 
approximate regularities) 
2.3 The Lexicon Learning system, L 2 
Previous experiments in automatic lexical acquisti- 
lion at. S1CS (L ~ - Lexicon Learning) used a set of 
1 The system does not attempt to cope with c|oaed-categc)ry 
words. '\['hey have to be entered into a apecific function-word 
lexicon by a skilled linguist. 
ACTES DE COLING-92, NANTES, 23-28 AO~r 1992 1 1 7 3 I'gOC. OF COLING-92, NAN'rES, AUG. 23-28, 1992 
sentences and a formal grammar to infer the lexi- 
cal categorit.'-s of the words in the sentences. The 
original idea wa.q to start with an empty lexicon, as- 
suming that the grammar would place restrictions on 
the words in the sentences sufficient to determine an 
assignment of lexical categories to them \[Rayner el 
al 88\]. This can I)e viewed as solving a set of equa- 
tions where the words are variables that are Io be 
assigned lexical categories and the constraints that 
all sentences parse with respect to the grammar are 
the equations. 
Unfortunately, it proved almost impossit,le to 
parse sentenees containing several nnknown words. 
For this reason the scheme was revised in several 
ways \[tlgrmander 88\]; instead of starting with an 
eu/pty lexicon, the starting point bccanw, a lexi- 
con coutaining clnsed-cl;kss words snell ;L~ l)FOllOIlnS~ 
prepositions and determiners. The system would 
then at each stage only process sentences that coil 
rained exactly one unknown word, the hop,, I)eing 
that tlie words learned from these sentences would 
reduce the number of unknown words in the other 
ones. In addition to this, a rnorphologicat component 
w~s included to guide the assignments. Although the 
project proved the femsibility of the scheme, it also 
revealed some of its inherent problems, especially the 
need for fa.ster parsing methods. 
2.4 Explanation-based learning, EBL 
A problem with all natural language grammars is 
that they allow a vemt number of possible con- 
structions that very rarely, if ever, occur in real 
sentences. The application of explanation-based 
learning ~ (EBL) to natural language processing al- 
lows us to reduce tim set of possible analyses and 
provides a solution to the parsing inefficiency prob- 
lem mentioned above (Subsection 2.3). 
The original idea \[Rayner 88\] was t.o bypass llOl'- 
lna\] processing and instead use a set of learlled rules 
that perh)rmed the t.~qks of the normal parsing com- 
ponent, l:ly indexing the learned rules eflicieutly, 
analysing an input sentence using the learned rules is 
w~ry much faster than normal processing \[Samuels- 
son & Rayner 9t\]. The learned rules can be viewed 
as templates for grammatically correct phrases which 
are extracted from the. granmmr and a set of training 
sentences using explanatiou-bmqed learning, llere, we 
assume the following definition: 
Definition 2 
a ten'tplate ts a generalization constrvcted from lhe 
parse tree for a successfidly processed phrase, .,1 tem- 
plate is a tree spanning the parse with a mother cat- 
egory as root and a collection of its ancestor nodes 
2t~xplanation-lmsed learning is n machine learning tech- 
Illqlle closely related to tllaCro-operator learllil|g, chtlllkillg, 
and parliM evaluation and is described in e.g.. \[I)e.long & 
Mooney 8~';, Mitchell et at 86\]. 
(at arbitrary, but pre-defined, deep levels of nesting) 
as I~a~les. 
The fact that the templates are derived from the 
original gramlnar guarantees that they represent cor- 
rect phrlLses and the fact that they are extracted from 
real senteuces ensnres Ihat they represent construc- 
tions that actually occur. 
3 Explanation-based 
lexical learning, EBL 2 
The basic algorithm goes ,xs follows: 
1. Using a large corpus from the domain, extract 
teUll)lates from the sentences contaiuing uo 1.111- 
known words. 
2. Analyse the remaining sentences (the ones con- 
taiuing unknown words) using the templates, 
while maintaining an interim lexicon for the un- 
known words. 
3. Compare the restrictions placed on the unknown 
words by the analyses obtained with other hand- 
coded phrase templates specific for the para- 
digms m the lexicon 
d. (2reate "rear' lexical entries from the mforma- 
ti<m m the intcrhn lcxicon when a full set of 
such templates \[covering a paradigm) has been 
found. 
In the following subsections, we will address these 
issues in turn. 
3.1 Extracting templates from 
a domain-specific corpus 
A typical situation where we think that this method 
is well suited is when a general purpose NL system 
with a core lexicon (such as the S-CLE) is to be cus- 
tomized to a specific application domain. Tile vocab- 
ulary used in the domain will include e.g. technical 
terms that are not present in the core lexicon. Also, 
the use of the words in the core lexicon may differ 
between domains. In addition to this, some types 
of gralnmatieal coustrilcts may be more eonllnon ill 
one domain than ill allother. We will try to "get the 
flavour of the language" in a particular application 
euviromnenl from domain-specific texts. 
The corpus is divided into two parts: one with 
seatellces containing ilnknown words, all(\] another 
where all the words are known, The latter group 
is used to extract plmme templates that capture 
tile grammatical constructions occurring in tile do- 
main. rFhe process of extracting phrase templates 
from training sentences is outlined in Subsection 2.4. 
AcrEs nl~ COTING-92, NAt,rl~s, 23-28 Ao(rr 1992 1 ! 7 4 PRec. OF COLING-92, NAmV:s, AUG. 23-28, 1992 
3.2 Analysing the 
remaining sentences 
Assuming that a partieular set of phrase tenlplales 
is applicable to a sentence containing an unknown 
word will associate a set of constraints with the word. 
Naturally, the constraints Oil I\[le kBowlt words of 
the sentence should be satisfied if this tcmplatv is 
to be e(msidered. 3 This will correspond to a partic+- 
ular parse or analysis of the seutenee. Thus a sol of 
constraints is a.ssociated with each different pm'se 
A number of entries in the prot.otype iexicou will 
matcll the set of constraints associated with a sen- 
teuce. \['\]aeh prototyI)e is all illCal'llatioIl of il para 
digtn, Thus we can a.ssociate a word with a set of 
paradigms. (Note thai the paradigms may be non- 
exclusive.) All such +msociatious (corresponding to 
different parses of the same sentence) are collected, 
and used to update the+ interim h'xieon. 
'\['h(! IllOSt obvious conslraiuts colnt! frolll syllt{ic 
tie considerations. If, in Ihe sentence John loves a ca( 
the word loves were unknown, while the other words 
did indeed have the obvious lexicai entries, the gram 
mar will require loves to be a transitive verb of third 
person singular agreement. Since the prototYl)eS of 
verbs are iu tl,e imperative form, we nmst associate 
a finite verb form with the imperatiw~, This is done 
by applying a omrphologieal rule that strips the '-s' 
from the word loves, reinforcing the hypothesis and 
gaining the tense information in the process. 
Now, this ntorphological information lnay seem 
uniml)ortant in Fnglish, but it definitely is +lol it, 
Swedish: a word with more that+ one sy\]lal,h+ end- 
ing with '-or' has to be an in(h.finite common gel,der 
noun. If it is not of latin origin it lnusl, be a phi 
ral form an(I thus ils entire morl)hology is kJvm, n 
The odds that it is a countabh" noun (like d.ck), as 
(}\[)posed tO 1t lllaSS IIOIln (such {IS walev), ;ll'C ()vet" 
whehning. 
3.3 Constructing lexical entries 
During tile analysis of the set of sentences conlain- 
ing unknown words, an interim lexicon for these un- 
known words is kept. The interim lexieon is imlexed 
on word sterns and updated each titlie a IWW Sell 
fence is i)roeessed. \["or each word sI, eul+ t'e/o pieces 
of information are retained in this lexicon: a hypo- 
thesis about which paradigm or set. of paradigms lhe 
word is assumed to belong to, and a justifieat.ion Ihat 
encodes all evidence relevant to the word. The jnsti- 
fieation is used to make the hypothesis aml is main 
tained so that the entry may be Ul)(lat, ed whett new 
inlbrmation about tim word arrives. When all the 
l)hrase templates (sentence patterns) for lhlfilhnent 
3 UldeSs tile)' Ih) ill fact COll't!sp(lltd to othtT llr)ll lexicaliz,:d 
Sl¢llSeS of tile word, in' to hO|llO~l.&l)hS, 
of a Sl)ecilic para(ligm have been found, an entry for 
the word is made in the domaimspecifie lexicon that 
is bcmg constructed. This is done while still keeping 
the justilication reformation, since this might con- 
taht evidence indicating other word-senses or holno- 
graphs 
4 hnI)lementation status 
A prelimiuary versi(~u of the lexieal acquisition sys 
tern has been implemented in Prolog. "File meal- 
tile extracting telnplates froln Selltences with knowll 
words is \[uily operational. The parser for sentences 
witil unkuown words has also been tested, while tile 
iaterim lexicon still is subject to experimentatiolL 
Presenl.ly, a w'ry siml)le strategy for the interiln lex- 
icon has been tesled. This version uses the set of 
all hypotheses ns the justification and use their dis- 
.itmetion as the era'rent hypothesis. We are currently 
working Oll extending this sdlenle to one incorporat- 
ing the full algorithm deseril)ed above. 
Unknowu wor(l~ are matched with tim subcalego- 
rizatiou paradigms of the S-CLE. In total 62 differ- 
enl synl.aet.ic/semantic paradigms are known by the 
present systmn: 5 for Swedish nmms, l0 for adjec- 
tives, aud all tim others for verbs. Tim morphologi- 
cal inflections are subdivided into 14 different inflec- 
tional cbLsses of nouns, 3 classes of adjectives, and 24 
classes of verbs. 
5 Conclusions 
We have (mt.lin<'d a method for autonlatic lexieal ae- 
(luisilion. An existing lexicon built on the usage 
of i)rolotypica\] entries for l)aradigms of opemela.ss 
words, is ext.end~'d b 5 infi~rring new lexical entries 
fl'OIII texts containing Dnkl/own words. The COll- 
straints placed on these words by the gramnlar arc 
compared with the prototypes and a hypothesis is 
made al)ouI what paradigm the word is most likely 
to l)olong to. 
The hy\]lotheses ai)otlt, the ilnknown words are 
kept+ m an interim lexicon until a suflicient level of 
confidence is reached. Phrase templat<~s are both 
hand-cod<+d aud extracted front the grammar and 
donlaiu-spt!citic texts using an explanation-based 
h,arning method. 
6 Acknowledgements 
The work reported here was fimded by the Founda- 
tion tot the Swedish Institute of Computer Science 
aud the Swedish National Board for Industrial and 
T<,ch nical l)evelol)mel\]t ( N UTEK). 
Aeries nE COLINGO2. N,~t~"nis, 23-28 ^ot~rl 1992 1 l 7 5 I)roc. OF COLlNG-92, NANTES, AU(). 23-28, 1992 
We would like to thank Manny Rayner and David 
Carter (SRI Cambridge) and Seif llaridi (SICS) for 
helpful discussions and suggestions, and Pierre Gan- 
der (Stockholm University) for valuable supl)ort. 
References 
Alshawi, 11. and J. van Eijek (1989). "Logical t"orms 
in the Core Language I';ngine", the 271h An- 
nual Meeling of the Association for Coalpala- 
tional Linguistics, Vancouver, llritish Columbia, 
pp. 25- 32. 
Alshawi, tl., D. ('.after, B. Gaml)iiek and M. Ray- 
net 11991 ). "Translation by Quasi Logical Form 
Transfer", the ~9lh Annual Meeting of th~ Asso- 
ciation for Computational Li)tgaisltes, Univer- 
sity of California, Berkeley. California, pp. 161.. 
168. 
Alshawi, I1. (ed.) (1992). 7'h~ Core Langttage 
Engine, C'ambridge. Massachusetts: The MIT 
Press. 
Boguraev, B., T. tlriscoe, J. Carroll, D. Carter 
and C. Grovcr (1987). "The Derivation of a 
Grammatically Indexed Lexicon from the Long- 
lnan Dictionary of Contemporary English", the 
251h Annual Meeting of the Associattoa for 
Computational Linguistics, Stanford, Califor- 
nia, pp. 193-200. 
C.arter, D. 11989). "Lexical Acquisition in the (_lore 
Language Engine", the 4th Conference of the 
European Chapter of the Association Jbr Com- 
putational Liaguisttcs, Manchester, Eugland, 
pp. 137- 144. Also available as SRI lnternat ioaa\[ 
Report CCSRC-012, Cambridge, England 
Calzolari, N. and R.. Bindi (1990). "Aequisilion 
of Lexical Information from a Large Textual 
Italian Corpus", Ihe 131h International Con- 
ference on Computational Linguistics, lh'lsinki, 
Finland, Vol. 3, pp. 54 59. 
DeJong, G and R. Mooney 11986). "Explanation 
Based Learning: An Alternative View". Ma- 
chine Learning, 1:145-- 176. 
Ganfl)~ck I:L and M. ll.ayner (1992). "The Swedish 
Core Language Engine", the 2rd Nordic Confer- 
ence of Tex! Comprehension in Man aad Ma- 
chine, LinkSping, Sweden. 
Gamb/iek B. (1992). "I,exieal Acquisition: The 
Swedish VEX System", the 2rd Nordic Confer- 
ence of Tert Comprehension in Man and Ma- 
chine. LinkiSpiag, Sweden. 
Granger, R.It. (1977). "FOlrl,-UP: A program that 
figures Otlt meallings of words from colltex(', 
the 5tb lnter~*atio)ml Joint Conference on Ar- 
tificial Intelligence, Cambridge, Massachusetts, 
pp. 172- 178. 
Grosz, B.J., D.E. Appelt, P. Martin, and 
F.C.N. Pereira 11!)87). "TEAM: An Experi- 
ment in the Design of Transportable Natural- 
Language Interfaces", Artificial Intelligence, 
32:173 243. 
tl6rmander, S. 11988). "The Problems of Learning 
a 1,exicou with a Formal Grammar", SICS R,e- 
search l/eport R88019, Stockholm, Sweden. 
Jacob.~. P. a~l¢l U. Zernik 11988). "Acquiring Lexieal 
Knowledge from Text: A Case Study", the 7th 
National Conference on Artificial Intelligence, 
Saint Paul, Minnesota, pp. 739 744. 
Mitchell, T.M., ItM. Keller and S.T. Kedar-Cabelli 
(1986). "Explanation-ltased Generalization: A 
Unifying View". Machine Learning, 1:47 80. 
Rayner, M. 11988). "Applying Explanation-Based 
Generalization to Natural-Language Process- 
ing", the lntcrnalional Conference on Fifth 
Geaeration C'ompuler Systems, 'lk)kyo, Japan, 
pp. 1267- 1274 
Rayner, M., /~. Hugosson and G. Ilagert (1988). 
-lIsmg a Logic Grammar to Learn a Lexicon", 
the 12th International Conference on Computa- 
tional Linguistics Budapest, llungary, pp. 524- 
529 Also available ,as SICS Research Report - 
R88001, Stockhohu, Sweden. 
Samuelsson, C. and M. Rayner (1991). "Quantita- 
tive Evaluation of Explanation-Based Learning 
a~ an Optimization Tool for a Large-Scale Nat- 
ural i,anguage System", the 12th International 
Jotnl Conference on Artificial Intelligence, Syd- 
ney, Australia. 
Trost. tl. and E. Buchberger 11986). "Towards 
the Automatic Acquisition of Lexieal Data", 
the llth International Conference on Computa- 
tional Linguistics, ltonn, Germany, pp. 387 389. 
Wilensky, R. (1990). "Extending the Lexicon by 
Exploiling Subregnlarities", the DA RPA Speech 
and Natural Language Workshop, llidden Val- 
ley, Pennsylvania, pp. 365-370. 
Zernik, U. and M. Dyer 11985). '"lbwards a Self- 
Extending Lexicon", the 22rd Annual Meeting 
of the Associatio~t for Computational Linguis- 
tics, University of Chicago, Chicago, Illinois, 
pp 284 292. 
Zernik, U. (1987). "l:mguage Acquisition: Learning 
a lIierarehy of Phrases", the lOth International 
Joint Conference o)1 Artzficial lntelliqence, Mi- 
lan, italy, pp. 125 131 
ACTXS DE COLING-92, NAN'W.S, 23-28 AO(Yr 1992 1 I 7 6 PROC. OF COLING-92, NANTES, AUG. 23-28, 1992 
