IRList Digest Thursday, 11 December 1986 Volume 2 : Issue 70 Today's Topics: Query - Back issues on knowledge representation, AI applications Announcement - New relationship between SIGIR and IP&M NL-KR - NL dialogue in an integrated computational model CSLI - Role of lexical semantics in syntax, Resource bounded agent behavior COGSCI - Massively concurrent systems for KR and reasoning News addresses are ARPANET: fox%vt@csnet-relay.arpa BITNET: foxea@vtvax3.bitnet CSNET: fox@vt UUCPNET: seismo!vtisr1!irlistrq ---------------------------------------------------------------------- Date: 1 December 1986, 16:22:02 MEZ From: Wolfgang Zocher (0511) 762-3684 ZZZO at DHVRRZN1.bitnet RRZN Universitaet Hannover Schlosswender Str. 5 D3000 Hannover 1 Subject: back issues by topic Dear Ed, to get started, I need some back issues of the following topics: 1. Knowledge Representation 2. AI Applications Thanks, Wolfgang Zocher (zzzo@dhvrrzn1) [Note: I have no topical organization that would allow me to select issues easily to satisfy your request. I will be happy to supply desired issues by number; perhaps other readers would send you a list of numbers they think satisfy your needs and we could work from that. It might be hard to decide which articles are relevant to YOUR query, though. Sorry for not being able to help further, Ed.] ------------------------------ Date: Wed Dec 10 17:01 EST 1986 From: fox Subject: New Arrangement Between ACM SIGIR and IP&M I. ANNOUNCEMENT ACM SIGIR (Special Interest Group on Information Retrieval) has many members involved or interested in research activities. Some of those members receive news through IRList Digest, but only if they have access to a computer network reachable from BITNET or the DARPA Internet. All SIGIR members receive the Forum, which is an unrefereed newsletter. As a additional aid to SIGIR members, and to better serve the readers of "Information Processing and Management," an international journal published by Pergamon Press, a new arrangement has been established between ACM SIGIR and IP&M. 1) Pages Set Aside Since SIGIR members are actively involved in research that is of interest to IP&M readers, there will be a number of pages set aside each year for SIGIR use. Some of the possibilities are listed below; other suggestions are welcome. 2) SIGIR Meetings IP&M will publish selected papers from SIGIR meetings, and will if agreeable to all, publish a special issue where such papers are assembled. 3) Special Issues IP&M is interested in special issues on topics related to SIGIR members' work and/or braoder interests. An example of this is the special issue on "Artificial Intelligence and Information Retrieval" now being assembled by W. Bruce Croft. An editor can be appointed for other such issues when a suitable topic is identified. 4) Articles IP&M is interested in receiving other articles from SIGIR members. They should be sent to the Liason (see 7 below) and will be reviewed by at least 2 referees in standard fashion. Submissions to SIGIR Forum that have such potential will be screened with this in mind. 5) Camera Ready Publications Brief communications can appear quickly in IP&M if sent in camera ready form -- providing a fast publication outlet. 6) Announcments IP&M is willing to publish SIGIR announcements such as calls for papers, free of charge. 7) Liason Dr. Edward A. Fox has been appointed to serve on the editorial board of IP&M. He will serve as a liason between SIGIR and IP&M. He can be contacted to receive submissions of any of the types mentioned above, or for other related matters. Correspondence should be sent to Dept. of Computer Science, Virginia Tech, Blacksburg VA 24061 II. CALL FOR CONTRIBUTIONS In light of the above arrangements, a call is hereby made to all SIGIR members to send in articles, announcements, suggestions for special issues, or other comments relating to this new arrangement. Please communicate directly with the Liason and be sure to indicate that you wish your submission to be considered for IP&M. ------------------------------ Date: Sun, 7 Dec 86 01:09:47 est From: nl-kr-request@cs.rochester.edu Subject: NL-KR Digest Volume 1 No. 26 [Extract - Ed] Date: 1 Dec 86 14:42:53 EST From: Robert.Frederking@cad.cs.cmu.edu Subject: Thesis Defense: Frederking on Dialogue in a Computational Model NATURAL LANGUAGE DIALOGUE IN AN INTEGRATED COMPUTATIONAL MODEL Robert Eric Frederking ABSTRACT Natural language dialogue is a continuous, unified phenomenon. Speakers use their conversational context to simplify individual utterances through a number of linguistic devices, including ellipsis and definite references. Yet most computational systems for using natural language treat individual utterances as separate entities, and have distinctly separate processes for handling intersentential phenomena. The computational system presented here, Psli3, uses the uniform framework of a production system architecture to carry out the different processes involved in natural language understanding and generation in a well-integrated way. Its conversational context arises naturally as the result of the persistence of the internal representations of previous sentences in working memory. Natural language input is interpreted within this framework using a a modification of the syntactic technique of chart parsing, extended to include semantics, and adapted to the production system architecture. It provides a graceful way of handling ambiguity within this architecture, and allows various knowledge sources to act in a highly integrated fashion. The smooth integration of different types of processing across different utterances is demonstrated using the phenomena of intersentential ellipsis and definite reference. The design of this system demonstrates how flexible and natural user interactions can be carried out using a system with a naturally flexible control structure. It also demonstrates how one could implement a real natural language interface in a sufficiently fast production system. The semantic chart parser is further extended to allow several closely related sentences to be treated in a single chart. This allows the relationship between the sentences to be used in a simple way to select between competing alternative intrepretations. We describe this system in detail, and include a number of extensive examples of the system's processing during a user interaction. A processing-based taxonomy for ellipsis resolution that we developed is also presented. ------------------------------ Date: Sun, 7 Dec 86 01:09:09 est From: vtcs1::in% Subject: CSLI Calendar, December 4, No. 9 [Extract - Ed] Reading: What to do with theta-Roles? Discussion led by Annie Zaenen December 4 When Extended Standard Theory won the linguistic wars (Newmeyer's version of linguistic history), lexical semantics went out of fashion in mainstream generative grammar but, as is often the case with victories that are the results of power politics rather than reason, the problems that were raised in the generative semantics research remained unsolved and recent years have seen them resurface. At this point several attempts to specify the role of lexical semantics in syntax are under elaboration. Among the debated issues are (1) the way semantic information has to be represented in the lexicon; (2) the number and the properties of the levels of representation needed to link semantics and syntax. The paper tries to give a partial answer to these questions from a Government Binding related view. I choose it because that point of view will most likely not be widely represented among the live participants at the TINLunch. The main purpose of the lunch should be a discussion of the general issues raised in the paper rather than a critique of the paper itself. Other relevant recent writings on the topic include: Dowty (1986): On the semantic content of thematic roles; Jackendoff (1986): The status of Thematic Relations in Linguistic Theory; Foley and Van Valin (1984): Functional Syntax and Universal Grammar; and Kiparsky's manuscript on Morphosyntax. -------------- THIS WEEK'S SEMINAR Rational Behavior in Resource-bounded Agents David Israel December 4 Members of the Rational Agency Project at CSLI (RatAg) have been involved in research to develop an architecture for the production of rational behavior in resource-bounded agents. The overall aim of this work is to combine techniques that have been constructed in artificial intelligence for automating means-end reasoning with a computational instantiation of techniques that have been developed in decision theory for weighing alternative courses of action. The focus is on ensuring that the resulting synthesis is a viable architecture for agents who, like humans and robots, are resource-bounded, i.e., unable to perform arbitrarily large computations in constant time. Predicating the architecture on the fact that agents have resource bounds will enable its use both as a device for producing rational behavior in robots that are situated in dynamic, real-world environments, and as a model of human rational behavior. In taking seriously the problem of resource boundedness, we draw heavily on the view of plans as ``filters'' on practical reasoning. We are concerned with determining what regularities there are in the relationship between an agent and her environment that can be exploited in the design of the filtering process. ------------------------------ Date: Sun, 7 Dec 86 01:09:51 est From: DEJONG%OZ.AI.MIT.EDU@XX.LCS.MIT.EDU Subject: Cognitive Science Calendar [Extract - Ed] Date: Friday, 5 December 1986 11:09-EST From: JHC at OZ.AI.MIT.EDU Subject: Revolving Seminar December 11 Thursday, 11 December 4:00pm Room: NE43-8th floor playroom AI REVOLVING SEMINAR Massively Concurrent Systems For Knowledge Representation And Reasoning Gul A. Agha, MIT AI Lab The problem of reasoning is central to Artificial Intelligence systems. The effectiveness of a "reasoning method" is intimately tied to the "knowledge representation" scheme on which it operates. The seminar will discuss some recent theoretical work in inheritance-based models for knowledge representation. Problems germane to inheritance-based models include exception handling, multiple inheritance, and viewpoints. The talk will outline some mechanisms that have been proposed to address these issues. Methods of reasoning such as first-order logic, nonmonotonic logic and due process reasoning will be related to the knowledge representation schemes. ------------------------------ END OF IRList Digest ********************