This report summarizes the capabilities of five computer programs at Yale that do automatic natural language processing as of the end of 1976. For each program an introduction to its overall intent is given, followed by the input/output, a short discussion of the research underlying the program, and a prognosis for future development. The programs…

This research is designed to fulfill the need of scholars, teachers, and developers of teaching materials who are active in the study and compilation of Russian textbooks. Through computer processing, the idioms were classified on the basis of pertinent linguistic criteria, including phonological, morphological, and syntactic information. Included…

Computer programs for linguistic analysis of language samples from bilingual children were surveyed in order to evaluate their usefulness. Eight programs which could be implemented on the UCLA IBM 370/3033 computer were considered. It was determined that the Computer Assisted Language Analysis System was the most promising in terms of capabilities…

This report describes the current version of the Why System, a script-based Socratic tutor which uses tutoring strategies formulated as production rules. The current system is capable of carrying on a dialogue about the factors influencing rainfall by presenting different cases to the student, asking for predictions, probing for relevant factors,…

Model building techniques from Artifical Intelligence and Information-Processing Psychology are applied to human-computer interface tasks to evaluate existing interfaces and suggest new and better ones. The model is in the form of an augmented transition network (ATN) grammar which is built by applying grammar induction heuristics on a sequential…

An expectation based system, NGP, for parsing English noun groups into the Conceptual Dependency representation is described. The system is a part of English Language Interpreter (ELI) which is used as the front end to several natural language understanding systems and is capable of handling a wide range of sentences of considerable complexity.…

This report describes several of the design decisions underlying ACT, a production system model of human cognition. ACT can be considered a high level computer programming language as well as a theory of the cognitive mechanisms underlying human information processing. ACT design decisions were based on both psychological and artificial…

The Commission of the European Communities has acquired an automatic translation system (SYSTRAN), which has been put into operation on an experimental basis. The system covers translation of English into French and comprises a dictionary for food science and technology containing 25,000 words or inflections and 4,500 expressions. This report…

The product of researchers' efforts to develop a computer processor which distinguishes between relevant and irrelevant information in the database, Spreading Activation Processor for Information Encoded in Network Structures (SAPIENS) exhibits (1) context sensitivity, (2) efficiency, (3) decreasing activation over time, (4) summation of…

A computer natural language system called HWIM (Hear What I Mean) is described in this report. Noting that the system accepts either typed or spoken inputs and produces both typed and spoken responses, the report proposes HWIM as an example of a relatively complete language system illustrating how the many components of language processing…

Language teaching requires textbook material that contains the most frequent concepts of a language. The computer brings its tremendous information processing ability to the task of establishing word frequency rankings, but the computer is limited to counting word-forms and not semantic concepts. The most recent word frequency dictionaries, in…

The paper surveys the major projects on the understanding of natural language that fall within what may now be called the artificial intelligence paradigm of natural language systems. Some space is devoted to arguing that the paradigm is now a reality and different in significant respects from the generative paradigm of present-day linguistics.…

Abstracts of research papers in computer and information science are given for 68 papers in the areas of information storage and retrieval; human information processing; information analysis; linguistic analysis; artificial intelligence; information processes in physical, biological, and social systems; mathematical techniques; systems…

The report documents performance on a 24-month R&D effort oriented toward the development of a computerized model for machine translation of natural languages. The model is built around a set of procedures called verbalization, intended to stimulate the processes employed by a speaker or writer in turning stored information into words.…

This article provides a tutorial introduction to Artificial Intelligence (AI) research for those involved in Computer Assisted Instruction (CAI). The general theme is that much of the current work in AI, particularly in the areas of natural language understanding systems, rule induction, programming languages, and socratic systems, has important…