58 HAM-ANS is being implemented in LISP/FUZZY on the DECsystem 1070 (PDP-10) of the Fachbereich Informatik at the University of Hamburg (cf. HOEPPNER 1981). This research is currently being supported by the Bundesminister fuer Forschung und Technologic (BMFT) under contract 081"1'15038. HAM-ANS (Hamburg Application-oriented Natu raI-Language System) U.of Hamburg, FRG Prof. Dr. Walther von Hahn Dr. Wolfgang Wahlster Dr. Wolfgang Hoeppner Research Unit for Information Science and Artificial Intelligence University of Hamburg Germanisches Seminar Mittelweg 179 D-2000 Hamburg 13 Fed. Rep. of Germany (This group is to be joined by four research associates and six research assistants.) The aim of this three-and-a-half-year project (7/1/81-12/31/84) is the development of a robust and cooperative system which provides natural language access in German to other software systems, e.g. reservation, data base, and vision systems. The basic core of HAMANS is a generalized version of the experimental NL-system HAMRPM (see below). Broad coverage will mainly be achieved in the analysis grammar, the inference component and the language generation component. Domain-dependent parts are to be identified and implemented in a way that allows the exchange of different application areas while keeping up the core modules. The implementation will be accompanied by optimization efforts, e.g. compilation of LISP and ATN code. Issues of adaptability and portability of the interface will be studied by implementing parts of the system on a LISP machine and in FRANZ LISP on a VAX 11/780. Case studies are planned for the following task domains: an interface to a large relational data base system (dealing with fishery data and implementcd in PASCAL/R), an interface to a motion analysis system (cf. MARBURGER/NEUMANN/NOVAK 1981) and the hotel reservation system which was started in the HAM-RPM project (see above). In addition, the basic research in HAM-ANS will center around the following topics: Knowledge acquisition through natural language dialogues Maintaining and using an explicit partner model (cf. MORIK 1981) Analysis and generation of word formations (cf. HOEPPNER 1980) Explaining inference-based answers and speech acts (el. WAHLSTER 1981) Bibliographic references: W. HOEPPNER: Derivative Word Formation in Contemporary German and its Algorithmic Analysis. (Tuebinger Beitraege zur Linguistik 128), Tuebinger: Narr 1980 (in German). W. HOEPPNER: An Overview of the FUZZY programming environment and File Access in HAM-ANS. Memo ANS-1, September 1981 (in German). H.MARBURGER, B. NEUMANN, H-J. NOVAK: Natural'Language Dialogue about Motion iin an Automatically Analysed Traffic Scene. In: Proc. of the 7th IJCAI, Vancouver 1981. P. 49-51. K. MORIK: Processing of external and internal situations in belief systems. In: J. Siekmann (ed.): GWAI-81. German Workshop of Artificial Intelligence, Bad Honnef, January 1981. (Informatik Fachberichte 47) Berlin, Heidelberg, New York: Springer 1981. P 287-296 (in German). W. Wahlster: Natural Language Argumentation in Dialogue Systems. AI MethOds for the Reconstruction and Explanation of Approximate Inferences. (Informatik Fachberichte 48) Berlin, Heidelberg, New York: Springer 1981 (in German). HAM-RPM (Ham bu rge r Redepa rtne rmodell) U. of Hamburg, FRG Prof. Dr. Walther von Hahn Dr. Wolfgang Hoeppner Anthony Jameson, B.A. Dr. Wolfgang Wahlstcr University of Hamburg Germanisches Seminar Von-Melle-Park 6 D-2000 Hamburg 13 Fed. Rep. of Germany The Natural Language system HAM-RPM simulates a dialogue partner conversing about one of several interchangeable scenes in colloquial German. The overall goals of the research have been twofold: simulating natural language dialogue behaviour by means ofa.

/lp/association-for-computing-machinery/ham-ans-hamburg-application-oriented-natural-language-system-u-of-0ADQC9MPn6