IRList Digest Thursday, 27 November 1986 Volume 2 : Issue 61 Today's Topics: Email - Why multiple copies of #57? - People at SMU (and comment on TR collection for IR) Query - References on NLP Techniques in IR (clarification) Reply - References and Information on Brain Hemispheres Discussion - Dissertation Abstracts for IR? - AIList handling of dissertation abstracts etc. Call for Papers - SIGIR87 Conf. and Possible Funds for Travel Seminar - Recent Research Advances in IR Abstracts - Last Year's ASLIB Conf. at Wadham College News addresses are ARPANET: fox%vt@csnet-relay.arpa BITNET: foxea@vtvax3.bitnet CSNET: fox@vt UUCPNET: seismo!vtisr1!irlistrq ---------------------------------------------------------------------- Date: Tue, 11 Nov 86 13:15 EST From: Thomas Kunselman Subject: Re: IRList Digest V2 For some reason I received 5 copies of this file, Vol 2 Issue 57. Just thought I would let you know in case the problem is on your end. Thanks, [Note: I sent only 1 to the distribution point, but you and I and a number of others received multiple copies. We have not been able to figure out the cause but it seems to not have happened since. I look forward to when Virginia Tech is on the Internet (hopefully only a matter of months now! Meanwhile, all I can do is apologize. - Ed] ------------------------------ Date: Sun, 16 Nov 86 00:32:45 est From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: People at SMU, TR accumulation for IRList .. As you know, the mailer was broken here and you took us off the irlist mailing list. Also, Dr. Korfhage and one of the graduate students working on information retrieval is now in Pittsburgh. ... Dr. Eich will be teaching information retrieval next semester and wants to see a copy or two to see if it would help her class. .. I am still accumulating IR related materials from AILIST and TRLIST for you. ------------------------------ Date: Sat, 15 Nov 86 01:27:19 est From: guindon%milano.sw.mcc.com@MCC.COM Subject: Query: Text processing techniques and document retrieval I would like any references on text processing techniques (parsing, discourse analysis, etc.) that have been applied for the purposes of document/information retrieval. I am not interested in natural language interfaces per se, but rather in the processing of the documents to be retrieved. If such a bibliography does not exist, I would volunteer building one from the received references. Thanks Raymonde Guindon [Note: Thanks for clarifying the request that you sent earlier, which appeared in IRList V2 #58. I hope that people will send you info on their work, and look forward to receiving a bibliography from you for IRList and possibly for SIGIR Forum. - Ed] ------------------------------ Date: 12 Nov 86 20:57:42 PST (Wednesday) Subject: re: Brain Hemispheres From: "Earle_Kyle.WBST102a"@Xerox.COM In response to the request for info on this (see below) I will be sending you a few items from my extensive files on the general subject of brain research in separate packages for ease of tossing away as you see fit. Would appreciate it, though, if you would forward all to the requestor as I couldn't figure out how to get through. I don't think any of this really belongs on this dl so wouldn't advise broadcasting it in your digest. Earle Kyle Xerox Corp. 102-19A Webster, NY 14580 (716)422-3133 kyle.wbst@xerox.com [Note: the request mentioned was in V2 #57. I am sending the forwarded messages to ocbno@utkvx4.bitnet, who might want to send us all a (short) summary of what relates to Information Science. Thanks Earle! - Ed] ------------------------------ Date: Sat, 15 Nov 86 01:29:15 est From: mkh%sevax.prg.oxford.ac.uk@CS.UCL.AC.UK Subject: Six character limit in indxbib In IRList Digest V2 #53, Mitchell Wyle writes: For starters, I shall use Unix's addbib, sortbib, roffbib, indxbib, lookbib suite of programs. The manuals say that one can change the options of indxbib when it stems, stop lists etc. I have found the list of the 100 most common words (/usr/lib/refer/eign), but I can't figure out how to change the stemming from 6 characters. Can someone out their in IRdigest-land help? [Note: Perhaps Mike Lesk will read this or someone else at Bellcore will prod him to ask about this? - Ed] Any information on this would be very welcome here too - we are currently using indxbib/lookbib etc. and the lower-level utilities they call as the basis of our (small and rudimentary) online catalogue, but are having problems with false drops caused by this 6 character limit. David Brown, Librarian, Oxford University Computing Laboratory ------------------------------ Date: Wed, 22 Oct 86 23:10 EDT From: KROVETZ%cs.umass.edu@CSNET-RELAY.ARPA To: ailist-request@SRI-STRIPE.ARPA Subject: AI-related dissertations [Note: this was forwarded by Ken Laws; see next msg. - Ed] Ken, Laurence Leff asked a question in a recent AIList about the SIGART feature on AI-related dissertations. I compile that list (along with a former co-worker, Susanne Humphrey). I think the information could be forwarded to AIList if you'd like me to. I spoke to the person at Dissertation Abstracts International some time ago about making the list available electronically, and she didn't think it would be a problem, but said that I would have to write a letter asking for formal permission. SIGART just contains notices of the dissertations; Computational Linguistics gets a stripped down version with just the stuff related to natural language plus some stuff on knowledge representation, but they also publish the abstracts. If you'd like the information sent to AIList, would you want just the notices, or the abstracts as well? Considering the distribution that AIList gets, I think the abstracts might be too much, but then again there's a fair amount of volume in the bibliographies that get sent out. On a side note: I've been trying to pursue a pet idea I've had for some time - establishing a network library for computer science tech reports and dissertations. I've spoken with people at the CSNET information center, and they thought it was a great idea, but the problem is how to get it funded. What I have in mind is something like the info-server that already exists. I've spoken with people at the defense data network and NTIS, but couldn't get anywhere. I'm trying to pursue avenues at NSF, but unfortunately it doesn't fit neatly into any of their "designated categories", and the group that deals with miscellaneous stuff was recently eliminated. I'm still going to keep trying, but any suggestions would be appreciated. If it ever gets established, I hope to use it for some post-doctoral work (I want to do my dissertation on conceptual information retrieval). -bob krovetz@umass.csnet ------------------------------ Date: Fri, 24 Oct 86 02:17:46 edt From: LAWS@SRI-STRIPE.ARPA To: krovetz%cs.umass.edu@relay.cs.net Subject: Dissertation Notices [Note: The following message was sent along with the one above to me and I feel they are both of interest to readers, but are easiest to follow in the order given here. - Ed] I'm forwarding your message to Lawrence Leff and Ed Fox since they may have some comments. Thanks for your offer. I agree that the abstracts would be too much for AIList to publish, although I'm sure there are some readers who would like as much text as possible (if only for their keyword searches). The problem is that AIList is a broadcast medium, and it doesn't make sense to broadcast more than people are willing to read. The dissertation notices would be welcome. I do have negative feelings about stealing 10% of SIGART's content, but I suppose the audiences and archival properties are different. Certainly the retrieval properties are different. I leave it up you whether you want to submit the notices; if you do, I will forward them. Alternatively, you might submit them to Lawrence and he could incorporate them in his bibliography. I also have a desire to reduce the volume of AIList, which is taking so much of my time that I can't afford to send for and read any of the material appearing in these bibliographies. Everyone seems reasonably happy with the status quo, but if we were all paying for the network services and the man-hours spent reading this material, I wonder how many of us would ante up. The problem, again, is the broadcast nature of the list. There really should be a database server or perhaps umpteen sublists (as on Usenet) rather than one linear stream. AIList does serve as a useful forum for the interaction of many disciplines, but it really wouldn't harm anyone if there were separate lists for seminars, conferences, and bibliographies. A good database server would obviate the need to broadcast such messages at all, thus saving a lot of Arpanet bandwidth. Lawrence has already been building a database of CS reports. I'm not sure of the details, but he's definitely the man to talk to. Ed Fox is plugged into the information retrieval community and has an AIList retrieval system. Feel free to ask the list readership about these matters if you want more feedback. -- Ken Laws [Note: I have contacted the parties involved and asked for appropriate permissions so that we can publish dissertation information in IRList and SIGIR Forum. I welcome comments by readers on this topic. I too feel that there is need to support network distribution of information in a more orderly way than now must take place - but I guess we are all in the middle of an evolutionary phase of the transition to electronic publishing. - Ed] ------------------------------ Date: Wed, 12 Nov 86 15:41:10 cst From: Don Subject: Re: news for Europeans about New Orleans? There is a reasonable possibility, albeit far from definite yet, that we will be getting some funds to support partial support for travel for non-U.S. scholars to attend the ACM/SIGIR International Conference on Research and Development in Information Retrieval, to be held at the Monteleone Hotel in New Orleans, Louisiana, USA, on June 3-5, 1987. Extended abstracts or papers are due on December 15, 1986, as stated in the Call for Papers, published previously in IRList. Anyone interested in being considered for a travel stipend should contact me as soon as possible, under the optimistic assumption that we will get funded. ------------------------------ Date: Sun, 16 Nov 86 13:33:17 est From: fox (Ed Fox) Subject: Graduate Seminar Series - talk by Dr. Abraham Bookstein Virginia Tech Department of Computer Science Graduate Seminar Series - 11/18/86 @ 4pm Recent Research Advances in Information Retrieval Dr. Abraham Bookstein Graduate Library School University of Chicago ------------------------------ Date: Tue, 28 Oct 86 18:08:50 -0100 From: Wyle Subject: More new references Here are some entries from an Aslib conference at Wadham College last year: %A T Addis %T Extended relational analysis: a design approach to knowledge-based systems %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %A D Parkinson %T Supercomputers and non-numeric processing %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %A D R McGregor %A J R Malone %T An architectural approach to advances in information retrieval %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %A M J Allen %A O S Harrison %T Word processing and information retrieval - some practical problems %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T Clustering and nearest neighbour searching %A F Murtagh %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T Experimenting with the automatic classification of books %A P G Enser %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T An analysis of ranking for free text retrieval systems %A N Teskey %Z Z Raznak %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T Interactive information retrieval: an artificial intelligence to deal with biographical data %A G P Zarri %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T A case system processor for the PRECIS indexing language %A P Hancox %A F Smith %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T Linguistic methods in information retrieval systems %A J Rouault %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T Design of a system for the online elucidation of natural language search statements %A V Aragon-Ramirez %A C D Paice %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T Problem descriptions and user models: developing an intelligent interface for document retrieval systems %A H M Brooks %A P J Daniels %A Belkin %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T Information retrieval aids in an online public access catalogue: automatic intelligent search sequencing %A N N Mitev %A S Walker %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T HEADS: A Cataloguing advisory system %A W J Black %A P Hargreaves %A P B Mayes %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom %T An architecture for integrating data, knowledge, and information bases %A D A Bell %B Informatics 8: Advances in intelligent retrieval %I Aslib, Information House %C 26-27 Boswell St, London WC1N3JZ %E M Rowbottom