Newsgroups: comp.ai.nat-lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!news.moneng.mei.com!uwm.edu!lll-winken.llnl.gov!koriel!wnoc-sfc-news!wnoc-tyo-news!aist-nara!wnoc-kyo-news!atrwide!atr-la!awb
From: awb@itl.atr.co.jp (Alan W Black)
Subject: Re: Morphological Analyzer for English
In-Reply-To: rungsawa@news.enst.fr's message of 9 Nov 1994 16:29:22 GMT
Message-ID: <AWB.94Nov15081956@as53.itl.atr.co.jp>
Sender: news@itl.atr.co.jp (USENET News System)
Nntp-Posting-Host: as53
Organization: ATR Interpreting Telecommunications Research Labs.,Japan
References: <1994Nov9.135415.39201@ucl.ac.uk> <39qtd2$63o@enst.enst.fr>
Date: Mon, 14 Nov 1994 23:19:56 GMT
Lines: 89

In article <39qtd2$63o@enst.enst.fr> rungsawa@news.enst.fr (Arnon Rungsawang) writes:


   From: rungsawa@news.enst.fr (Arnon Rungsawang)
   In article <1994Nov9.135415.39201@ucl.ac.uk>, ucjtrja@ucl.ac.uk (Dr James Au-Yeung) writes:
   |> I wonder if there is any good morphological analyzer around for English.
   |> 
   |> Please email me at: j.au-yeung@psychol.ucl.ac.uk
   |> 
   |> Thanks.
   |> 
   |> james
   |> 

   Please let me known about this also, my email is rungsawa@inf.enst.fr
   Thank a lot.
   Arnon.

The Alvey Morphological Analyser is available by anonymous ftp.  Its
quite big but contains a comprehensive morphological system.  An English
dictionary of about 60,000 stems is included. The system is written
in Common Lisp.  A much larger dictionary of 60,000 stems is available
as part of the Alvey Natural Language Tools (jac@cl.cam.ac.uk for
details) but requires payment and a licence.

The flyer for the system is below including ftp directions

* Alan W Black ---  ATR Interpreting Telecommunications Laboratories *
2-2 Hikaridai                         email: awb@itl.atr.co.jp
Seika-cho, Soraku-gun,                tel: (+81) 7749 5 1314
Kyoto 619-02, Japan                   fax: (+81) 7749 5 1308

--------

The Edinburgh/Cambridge Morphological Analyser System (3.1)

The Morphological Analyser and Dictionary project (MAP) was a three
year project (from 1984-1987) undertaken at Universities of Edinburgh
and Cambridge by Graeme Ritchie and Alan W Black (at Edinburgh) and
Stephen Pulman and Graham Russell (at Cambridge).  It was part of the
wider Alvey Tools project to develop general natural language tools,
with reasonable theoretical basis, suitable for use by other
researchers and developers in the field.

When building a lexicon for natural language processing it seems
wasteful to have to list all forms of words (e.g. "walk", "walking",
"walked", "walks" etc.).  A more useful lexicon is one where
generalisations about words can be captured.  Thus instead of words our
lexicon contains `morphemes'.  `Morphemes' are user specifiable atomic
word parts, (typically roots and affixes).  Also the system allows
rules to be specified defining how these morphemes can be combined
to form words.

The MAP system offers the following facilities:
  -- a rule system for describing orthographic changes that
     occur at morpheme boundaries (based on the Koskenniemi
     two-level model).  For example 'e' insertion in joining
     the suffix '+s' to nouns (for plurals) or verbs (for third
     person singular), as is "boxes", "glasses" etc.
  -- a rule system for describing which morphemes may combine to
     form words (described as a feature grammar).  For example that
     the suffix '+ly' makes adverbs from adjectives, and "+er" forms
     nouns from a particular class of verbs.
  -- lexical rules for expanding and manipulating basic lexical
     entries.
  -- An interactive environment for compiling rule systems and
     debugging them.
  -- program level interfaces for incorporating the analyser and
     dictionary system into other programs.

The whole system is written in Common Lisp and has been tested in
a number of implementations.  Full documentation both for the user
and programmer is available.  Two moderately sized lexicons and
rules for morphological analysis of English are included.

The system is designed to be stand alone, allowing morphological
analysers to be developed.  It may also be used simply as a lexicon
system within a larger natural language processing system.  The system
is also available within the full Alvey Natural Tools version 3.0
integrated with a GPSG parser, and development environment including a
very large lexicon and wide coverage grammar.  The theoretical aspects
of the morphological analyser and dictionary system itself are described
in more detail in the book "Computational Morphology" (Ritchie et. al.
MIT Press 1992).

The MAP system is available free without licence, by anonymous ftp from
scott.cogsci.ed.ac.uk [129.215.144.3]:/pub/phonology/tools/MAP/MAP3.1.tar.Z
Further enquiries may be sent to Alan W Black (awb@ed.ac.uk).

