FOXEA@VTCC1 (07/22/88)
IRList Digest Thursday, 21 July 1988 Volume 4 : Issue 39
Today's Topics:
Query - Automatic abstracting
- Pedagogical model for IRS
Discussion - Soundex algorithm (from AIList)
- Soundex code (from AIList)
Call for Papers - ACL European Chapter 1989 Conference
Announcement - Contents of special issue of IP&M
News addresses are
Internet or CSNET: fox@vtopus.cs.vt.edu or fox@fox.cs.vt.edu
BITNET: foxea@vtvax3.bitnet (soon will be foxea@vtcc1)
----------------------------------------------------------------------
From: Gareth Husk <gareth@computing.lancaster.ac.uk>
Date: Thu, 23 Jun 88 11:30:20 bst
Subject: Information Request on automatic abstracting
Newsgroups: comp.theory.info-retrieval
Organization: Department of Computing at Lancaster University, UK.
...
This is a request for information to help with my thesis on automatic
abstracting that I am in the process on completing.
Does any one have information, or know of a source, on the work of
professional abstractors.
I need to know:
i) The number of new academic papers per year.
ii) The rate at which abstractors have to work.
iii) The rate of increase of academic publications.
iv) The effects on the quality of abstracts of the increase of load.
v) Whether the abstract services are beginning to lag noticably
under the load.
Add to this any information that you may consider pertinent.
Please reply by e-mail, if other people require this information I
will forward a summary of responses.
Gareth Husk
--
" Nine weeks and counting..."
UUCP: ...!seismo!mcvax!ukc!dcl-cs!gareth
JANET: gareth@uk.ac.lancs.comp
------------------------------
Date: Mon, 18 Jul 88 21:05:24 +0300
From: The Computer in Education Research Lab <C37@TAUNOS>
Subject: pedagogical model for IRS
I am engaged in research and development of a pedagogical model
attached to an IRS, guiding pupils in highschools in searching
activities. I will be grateful for any information about research
concerning searching strategies, IRS users model, IRS in schools etc.
thanks, avigail oren
[Note: there are a number of articles on searching strategies
(Marcia Bates had article in JASIS for example) and Penny Daniels
has just finished dissertation on user models (does anyone have
current email or complete postal address for her?). There is
a lot more - hope people will send you some and you will give
us a bibliography - Ed]
------------------------------
Date: 11 Jul 88 18:21:16 GMT
From: sundc!netxcom!sdutcher@seismo.css.gov (Sylvia Dutcher)
Subject: Re: Soundex algorithm
[Forwarded from AIList Digest Monday, 18 Jul 1988 Volume 8: Issue 15]
[Note: While I do not generally overlap with AIList, I thought it
worthwhile to let IRList readers know about the discussion below so
they can join in the discussion there, or so we can have a parallel
discussion on IRList. It would be nice for AIList readers to learn
about the various other, often better algorithms. - Ed.]
In article <12520@sunybcs.UUCP> stewart@sunybcs.UUCP (Norman R. Stewart) writes:
>
> The source I've used for Soundex (developed by the
>Remington Rand Corp., I believe), is
>
> Huffman, Edna K. (1972) Medical Record Management.
> Berwyn, Illonois: Physicians' Record Company.
I've written a soundex program based on the rules in Knuth's _Searching_and_
Sorting. These are also the rules used at the National Archives to sort
census data. These rules differ slightly from the ones posted by Mr. Stewart.
If you don't need to match anyone else's soundex, the most important rule is
to be consistent. I will insert Knuth's rules below.
>The algorithm is very simple,
1. Retain the first letter of the name, and drop all occurrances of a, e, h,
i, o, u, w, y in other positions.
>1: Assign number values to all but the first letter of the
>word, using this table
> 1 - B P F V 2 - C S K G J Q X Z
> 3 - D T 4 - L
> 5 - M N 6 - R
> 7 - A E I O U W H Y
2. Assign number values as above, except for 7.
>2: Apply the following rules to produce a code of one letter and
> three numbers.
> A: The first letter of the word becomes the initial character
> in the code.
> B: When two or more letters from the same group occur together
> only the first is coded.
> C: If two letters from the same group are seperated by an H or
> a W, code only the first.
3. If two or more letters with the same code were adjacent in the original
name (begore step 1), omit all but the first.
> D: Group 7 letters are never coded (this does not include the
> first letter in the word, which is always coded).
4. Convert to the form "letter, digit, digit, digit" by adding trailing
zeros or dropping rightmost digits.
BTW according to the reference in Knuth's book, this algorithm was
developed by Margaret Odell and Robert Russell in 1922.
>Norman R. Stewart Jr. * How much more suffering is
>C.S. Grad - SUNYAB * caused by the thought of death
>internet: stewart@cs.buffalo.edu * than by death itself!
>bitnet: stewart@sunybcs.bitnet * Will Durant
--
Sylvia Dutcher * The likeliness of things
NetExpress Communications, Inc. * to go wrong is in direct
1953 Gallows Rd. * proportion to the urgency
Vienna, Va. 22180 * with which they shouldn't.
------------------------------
Date: Wed, 13 Jul 88 10:35:06 EDT
From: "William J. Joel" <JZEM%MARIST.BITNET@MITVMA.MIT.EDU>
Subject: Re: Soundex algorithm
[Forwarded from AIList Digest Monday, 18 Jul 1988 Volume 8: Issue 15]
[Note: While I do not generally overlap with AIList, I thought it
worthwhile to have an algorithm for those who teach or use Prolog
in their research - Ed]
/* The following is source code for a Soundex algorithm written in */
/* Waterloo Prolog. */
/* William J. Joel*/
/* Marist College */
/* Poughkeepsie, NY */
/* jzem@marist.bitnet */
key(a,-1).
key(b,1).
key(c,2).
key(d,3).
key(e,-1).
key(f,1).
key(g,2).
key(h,-2).
key(i,0).
key(j,2).
key(k,2).
key(l,4).
key(m,5).
key(n,5).
key(o,-1).
key(p,1).
key(q,2).
key(r,6).
key(s,2).
key(t,3).
key(u,-1).
key(v,1).
key(w,-3).
key(x,2).
key(y,-2).
key(z,2).
soundex(Name,Code)<-
string(Name,Code1) & write(Code1) &
soundex1(Code1,A.B.C.D.Rem) &
string(Code,A.B.C.D.nil).
soundex1(Head.Code1,Head.Code)<-
keycode(Head.Code1,Code2) & write(Code2) &
reduce(Code2,T.Code3) & write(T.Code3) &
eliminate(Code3,Code4) & write(Code4) &
append(Code4,0.0.0.nil,Code).
reduce(X.(-2).X.Rem,List)<-
reduce(X.Rem,List).
reduce(X.X.Rem,List)<-
reduce(X.Rem,List).
reduce(X.Y.Z.Rem,X.List)<-
^X==Z &
reduce(Y.Z.Rem,List).
reduce(X.Y.Rem,X.List)<-
^X==Y &
reduce(Y.Rem,List).
reduce(X.nil,X.nil).
reduce(nil,nil).
eliminate(X.Rem,List)<-
lt(X,0) &
eliminate(Rem,List).
eliminate(X.Rem,X.List)<-
gt(X,0) &
eliminate(Rem,List).
eliminate(nil,nil).
keycode(H.T,N.CodeList)<-
key(H,N) &
keycode(T,CodeList).
keycode(nil,nil).
append(Head.Tail,List,Head.NewList)<-
append(Tail,List,NewList).
append(nil,List,List).
------------------------------
Date: Thu, 23 Jun 88 17:46:32 EDT
From: walker_donald e <walker@FLASH.BELLCORE.COM>
Message-Id: <8806232146.AA07753@flash.bellcore.com>
Subject: ACL European Chapter Call for Papers
ACL European Chapter 1989
CALL FOR PAPERS
Fourth Conference of the European Chapter
of the Association for Computational Linguistics
10-12 April 1989
Centre for Computational Linguistics
University of Manchester Institute of Science & Technology
Manchester, England
This conference is the fourth in a series of biennial
conferences on computational linguistics sponsored by the
European Chapter of the Association for Computational
Linguistics. Previous conferences were held in Pisa (Sep-
tember 1983), Geneva (March 1985) and Copenhagen (April
1987). Although hosted by a regional chapter, these confer-
ences are global in scope and participation. The European
Chapter represents a major subset of the parent Association
for Computational Linguistics, and is in its seventh year.
The conference is open both to existing members and non-
members of the Association.
Papers are invited on all aspects of computational linguis-
tics, including but not limited to:
morphology
lexical semantics
computational models for the
analysis and generation of language
speech analysis and synthesis
computational lexicography and lexicology
syntax and semantics
discourse analysis
machine translation
computational aids to translation
natural language interfaces
knowledge representation and expert systems
computer-assisted language learning
Authors should send six copies of a 5- to 8-page double-
spaced summary to the Programme Committee at the following
address:
Harold Somers
Centre for Computational Linguistics
UMIST
PO Box 88
Manchester M60 1 QD
England
It is important that the summary should identify the new
ideas in the paper and indicate to what extent the work is
complete and to what extent it has been implemented. It
should contain sufficient information to allow the programme
committee to determine the scope of the work and its rela-
tion to relevant literature. The author's name and address
(including net address if possible) should be clearly indi-
cated, as well as one or two keywords indicating the general
subject matter of the paper.
Schedule: Summaries must be submitted by 1st October 1988.
Authors will be notified of acceptance by 15th December.
Camera-ready copy of final papers prepared in a double-
column format on model paper (which will be provided) must
be received by 28th February 1989, along with a signed copy-
right release statement. Papers not received by this date
will not be included in the Conference Proceedings, which
will be published in time for distribution to everyone
attending the conference.
The programme committee will be co-chaired by Harold Somers
(UMIST) and Mary McGee Wood (Manchester University), and
will include the following
Christian Boitet (Grenbole)
Laurence Danlos (Paris)
Gerald Gazdar (Sussex)
Jurgen Kunze (Berlin, DDR)
Michael Moortgat (Leiden)
Oliviero Stock (Trento)
Henry Thompson (Edinburgh)
Dan Tufis (Bucharest)
Local arrangements will also be handled by Somers and Wood.
Please await a further announcement in October for more
details.
Exhibits and demonstrations: A programme of exhibits and
demonstrations is planned. Anyone wishing to participate
should contact John McNaught at the above address. Book
exhibitors should contact Paul Bennett also at the above
address.
------------------------------
Date: Thu, 23 Jun 88 19:10:36 cdt
From: radecki@fergvax.unl.edu (Dr. Radecki)
Subject: Contents of special issue of IP&M
As the Guest Editor of the special issue of
Information Processing and Management
on "The Potential for Improvements in Commercial Document
Retrieval Systems," I would like to inform interested parties that it has
just been published as Number 3 of Volume 24, and is now available to both
subscribers and non-subscribers.
The contents of this special issue is as follows:
Bernard M. Fry Editorial: Robert Maxwell and information
Tefko Saracevic processing
Harold Borko
Tadeusz Radecki Trends in research on information retrieval--The
potential for improvements in conventional
Boolean retrieval systems
Peter Smit Information impediments to innovation of
Manfred Kochen on-line database vendors
William S. Cooper Getting beyond Boole
M. E. Maron Probabilistic design principles for conventional
and full-text retrieval systems
Edward A. Fox Practical enhanced Boolean retrieval:
Matthew B. Koll Experiences with the SMART and SIRE systems
G. Salton A simple blueprint for automatic Boolean query
processing
Tadeusz Radecki Probabilistic methods for ranking output
documents in conventional Boolean retrieval
systems
Jitender S. Deogun Integration of information retrieval and database
Vijay V. Raghavan management systems
Robert M. Losee Integrating Boolean queries in conjunctive
Abraham Bookstein normal form with probabilistic retrieval models
M. H. Heine A logic assistant for the database searcher
Danny P. Wallace Estimating effective display size in online
Bert R. Boyce retrieval systems
Donald H. Kraft
Michael D. Gordon The necessity for adaptation in modified
Boolean document retrieval systems
David C. Blair An extended relational document retrieval model
------------------------------
END OF IRList Digest
********************