Our purpose in studying natural language understanding in conjunction with problem solving is to bring together the constraints of what formal representation can actually be obtained with the question of what knowledge is required in order to solve a wide range of problems in a semantically rich domain. We believe that these issues cannot sensibly be tackled in isolation. In practical terms we have had the benefits of an increased awareness of common problems in both areas and a realisation that some of our techniques are applicable to both the control of inference and the control of parsing. Early work on solving mathematical problems stated in natural language was done by Bobrow (STUDENT - (i]) and Chamiak (CARPS - [5]). However the rudimentary parsing and simple semantic structures used by Bobrow and Charniak are inadequate for any but the easiest problems. Our intention has been to build on B/RG Chris This work was supported by SRC grant number 94493 and an SRC research studentship for Mellish.

Distributed systems are multiprocessor information processing systems which do not rely on the central shared memory for communication. The importance of distributed systems has been growing with the advent of "computer networks" of a wide spectrum: networks of geographically distributed computers at one end, and tightly coupled systems built with a large number of inexpensive physical processors at the other end. Both kinds of distributed system are made available by the rapid progress in the technology of large-scale integrated circuits. Yet little has been done in the research on semantics and programming methodologies for distributed information processing systems. Our main research goal is to understand and describe the behaviour of such distributed systems in seeking the maximum benefit of employing multiprocessor computation schemata. The contribution of such research to Artificial Intelligence is manifold. We advocate an approach to modelling intelligence in terms of cooperation and communication among knowledge-based problem-solving experts.

Knowledge about a particular type of ore deposit is encoded in a computational model representing observable geological features and the relative significance thereof.

A program called "AM" is described which carries on simple mathematics research, defining and studying new concepts under the guidance of a large body of heuristic rules. The 250 heuristics communicate via an agenda mechanism, a global priority queue of small tasks for the program to perform, and reasons why each task is plausible (for example, "Find generalizations of'primes', because'primes' turned out to be so useful a concept"). Each concept is represented as an active, structured knowledge module. One hundred very incomplete modules are initially supplied, each one corresponding to an elementary set-theoretic concept (for example, union). This provides a definite but immense space which AM begins to explore.

Successive utterances in coherent discourse refer to the same entities.

Human programmers seem to know a lot about programming. This suggests a way to try to build automatic programming systems: encode this knowledge in some machine-usable form. In order to test the viability of this approach, knowledge about elementary symbolic programming has been codified into a set of about four hundred detailed rules, and a system, called PECOS, has been built for applying these rules to the task of implementing abstract algorithms. The implementation techniques covered by the rules include the representation of mappings as tables, sets of pairs, property list markings, and inverted mappings, as well as several techniques for enumerating the elements of a collection. The generality of the rules is suggested by the variety of domains in which PECOS has successfully implemented abstract algorithms, including simple symbolic programming, sorting, graph theory, and even simple number theory. In each case, PECOS's knowledge of different techniques enabled the construction of several alternative implementations. In addition, the rules can be used to explain such programming tricks as the use of property list markings to perform an intersection of two linked lists in linear time. Extrapolating from PECOS's knowledge-based approach and from three other approaches to automatic programming (deductive, transformational, high level language), the future of automatic programming seems to involve a changing role for deduction and a range of positions on the generality-power spectrum.

... we began to program a computer understanding system thatwould attempt to process input texts. An item crucial to our ability to accomplishthis task was what we called a script. A script is a frequently repeated causalchain of events that describes a standard situation. In understanding, when it ispossible to notice that one of these standard event chains has been initiated,then it is possible to understand predictively. That is, if we know we are in arestaurant then we can understand where an "order" fits with what we justheard, who might be ordering what from whom, what preconditions (menu,sitting down) might have preceded the "order", and what is likely to happennext. All this information comes from the restaurant script.Hayes, J.E., D. Michie, and L. I. Mikulich (Eds.), Machine Intelligence 9, Ellis Horwood.

This approach leads to an infinite sequence of algorithms I. Surface Models

I will remain Ion8 and deeply indebted to Herb Simon, H. T. Kung, AI Newell, Jay

—A class of LISP programs that is analogous to the finite-state automata is defined, and an algorithm is given for constructing such programs from examples of their input-output behavior. It is shown that the algorithm has robust performance for a wide variety of inputs and that it converges to a solution on the basis of minimum input information.IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS, VOL. SMC-8, NO. 8,