Search

Science and technology use a large variety of symbols to represent physical properties, chemical formulas, and mathematical expressions. Since data centers which codify and evaluate physical properties need to use this conventional symbolism, it is recommended that they adopt the symbols and terminology specified by the various International Unions, both for manual operations and for the creation of machine-readable data bases. It is demonstrated that these conventional symbols can be produced by communications devices which are compatible with the international standard codes for information exchange. A set of characters suitable for representing scientific data and text is presented and proposed as an extension of the International Organization for Standardization(ISO) information interchange code. The use of this extended character code by computer oriented data centers at the National Bureau of Standards is described. In addition, the kinds of equipment needed for this level of performance and criteria for their selection are outlined. (Author)

This is the fourth in a series of four reports describing a study of 1,614 junior high school mathematics and English students and 69 of their teachers that was undertaken to discover the effects of different teaching behaviors on cognitive and affective student outcomes. This booklet is the working manual used for coder training and includes information on the general procedure and organization of coding, coding of academic response opportunities, types of response opportunity, subject matter checklists, student initiated questions and comments, classroom format, peer tutor identification, dyadic teacher-student contacts, student created contacts, teacher afforded contacts, behavior related contacts, student behaviors, and social contacts as well as a sample copy of the coding sheet. (TJ)

A LINGUISTIC PROCESSING SYSTEM, WHICH HAS BEEN DEVELOPED FOR CHINESE TO ENGLISH MACHINE TRANSLATION, UTILIZING THE CONTEXTUAL ASSOCIATIVE METHOD WAS DESCRIBED. THE REPORT INCLUDES (1) AN EXPLANATION OF THE LINGUISTIC PROCESSING SYSTEM, (2) MORPHOLOGICAL AND SYNTACTIC ANALYSES, AND (3) AN ANALYSIS OF ENGLISH INFLECTION. MACHINE APPLICABILITY WAS STRESSED THROUGHOUT, AND TESTING OF LINGUISTIC RULES BY MACHINE WAS INITIATED AND PROVED VALUABLE. THE CHINESE INPUT WAS PROCESSED BY CHICODER, A DEVICE USED TO ENCODE CHINESE CHARACTERS. (KL)

FMS/3 is a system for producing hard copy documentation at high speed from free format text and command input. The system was originally written in assembler language for a 12K IBM 360 model 20 using a high speed 1403 printer with the UCS-TN chain option (upper and lower case). Input was from an IBM 2560 Multi-function Card Machine. The model 20 version has been superceded by a revised implementation operational on all other models of the IBM System/360. Increased speed is gained by restricting performance to text formatting only, eliminating editing and text scanning functions, such as spelling checking, index generation, etc. This document serves as language specifications and users manual and was completely generated by FMS. It contains an introduction to the system, including the implementation and project history; a chapter on using FMS; an FMS beginners cookbook, and an explanation of Model 20 tape file management. [Related documents are LI 003610, LI 003611, LI 003645, LI 003647 and LI 003648.] (Author/SJ)

The object of this theoretical inquiry is to examine the controversial issue of a fully automatic high quality translation (FAHQT) in the light of past and projected advances in linguistic theory and hardware/software capability. This first volume of a two-volume report discusses the requirements of translation and aspects of human and machine translation. Techniques involved in machine translation, i.e., hardware, software, and linguistics, are covered. Contributions of linguistics to machine translation and pertinent recent work in linguistics are also considered. Views of specialists concerning machine translation are presented along with the conclusions and recommendations of the report. An appendix includes relevant papers contributed by specialists on the topic of machine translation. Papers are also included in the second volume of the study. For Volume 2, see FL 002 937. (VM)

The comprehensibility of materials compressed and then expanded by means of an electromechanical process was tested with 280 Army inductees divided into groups of high and low mental aptitude. Three short listening selections relating to military activities were subjected to compression and compression-expansion to produce seven versions. Data indicate that expanding previously compressed materials to restore the word rate to normal may restore the comprehension of the material to very near normal when the compression/expansion is limited to 40%. Present results substantiate findings that factors limiting the comprehensibility of rapid speech reside more with the inability of the listener to process rapid rates of speech than with the signal distortion produced by the equipment or compression process. (Author)

The report presents a progress in theoretical linguistics, descriptive linguistics, lexicography, and systems design in the development of a German-English Machine Translation System. Work in the theoretical group concentrated on intrasentenial disambiguation and on improving certain parts of the system to achieve greater economy in processing. The linguistic group was engaged in correcting and updating the existing German and English lexical data bases by assigning syntactic and semantic selection restrictions to lexical items. Work in the system group concentrated on the reduction of the size of the existing linguistics research system lexical data base without information loss, on the conversion of this data base to the linguistics research system script format, on the construction of supporting programs to expedite and facilitate the updating of the linguistics research system word lists, and on the construction of part of the linguistics research system grammar maintenance and systems programs. (Author)

This document lists over 100 research projects on various topics related to linguistics and language. The topics covered are foreign- and native-language instruction, relevant bibliographies, research in contemporary language, communication, textbooks, lexicology, lexicography, linguistics, computational linguistics, machine analysis of language, machine translation, phonetics, psycholinguistics, sociolinguistics, speech disorders, native language maintenance, language tests, technology and media, and translation science. The project descriptions include the director's name and address and details on project status, goal, collaborators, and funding. For the first volume, see ED 058 767. (VM)

This report deals with advances in linguistic analysis and programing. The theoretical theses for this work were presented in the first annual report. This second report concentrates on progress in the descriptive analysis of German and English, and on computer programs which have been developed during the year. Linguistic work during the period was directed at expanding the dictionaries for both German and English. The number of lexical items coded was increased, as were the features associated with each item. The coding involved complex problems of linguistic description, such as treatment of adverbs and extended forms of verbs. Since adverbs had not been adequately classified in any existing grammars of German or English, a new classificatory schema is being developed. The extended forms of German verbs have been treated in a recent monograph, which has served as the basis for the lexical analysis of these constructions within the German-English MT System. These extended forms are particularly difficult in having a small group of verbs used primarily to depict the verbal component and nouns combined with them to carry the semantic portion of the expression. (Author/RL)

This annotated bibliography lists 118 selected articles in the field of applied linguistics and language instruction which appeared in the journal "Modern Nyelvoktatas" from 1963 through 1970. The two categories containing the largest number of entries are: "Applied Linguistics/Linguistics" and "Pedagogy and Russian Language Teaching." Other categories include: (1) Language Teaching and Theory of Information, (2) Computer Translation, (3) Mathematical Linguistics, (4) Theory of Teaching Material--Modern Language Books, (5) Audio-visual Language Teaching, (6) Methodology--Programing--Intensity, (7) English Language Teaching, and (8) German Language Teaching. (RL)

Almost all of the published work on the philosophical question of mechanical intelligence has argued in favor of one or the other of the polarities of possible positions. It is however possible to take a position between these two extremes, those of the True Believer and the Infidel. While this agnostic position is not a strong position in the sense of having a good many logically compelling arguments in its favor, it gains its viability by virtue of the weaknesses in the polar alternatives. Although agnosticism concerning the possibility of artificial intelligence is believed to be a position held by many computer scientists, there has been no attempt to present this position in the literature. In an attempt to remedy the situation, this report presents the agnostic's position regarding artificial intelligence. (Author/NH)

Described is a project which produced the book, In Darkness, to explore the use of DOTSYS III (a braille computer translation program) and to demonstrate the use of BRAILLEMBOSS (a braille page printer designed to emboss braille at similar or faster rates than teletypes) as a means of producing Grade II braille output. Explained are successive production stages, including the creation of an input data file, translation, brailling, and the use of the Howe Press paper-tape driven, automatic sterotype. Included is a log reflecting the timing of actual operations during the initial experimental use of the DOTSYS-BRAILLEMBOSS system. Direct output of the project is said to include a BRAILLEMBOSS produced copy of In Darkness; a set of sterotype plates for press embossing of In Darkness; and a significant amount of experience with DOTSYS III installed in a commercial time-shared computer system, the BRAILLEMBOSS, and the automatic sterotype. It is recommended that additional projects be planned to develop better instructions for input typists, to test input instructions, to provide general cost data under simulated operating conditions, and to test a rebuilt automatic sterotype which would use modern solid state logic similar to that used in the BRAILLEMBOSS. (GW)

This document contains the reports summarizing the main discussion held during the March 1972 Computational Linguistics Conference. The first report, "Computational Linguistics and Linguistics," helps to establish definitions and an understanding of the scope of computational linguistics. "Integrated Computer Systems for Language" and "Computer-Oriented Grammars and Parsing" deal with technical issues and immediate concerns for research. "Machines and Speech" reports on a topic of emerging relevance in the field; lengthy consideration is given to reading machines for the blind and speech understanding systems, among other topics. "Language Performance (Psycholinguistics and Dialectology)" describes another fairly new application for computational linguistics. The last two reports, "Social Implications of Automatic Language Processing" and "Professional Ethics, Standards, and Education," involve issues related to the potential impact and implications of computational linguistics and its concerns. (VM)

This document lists over 50 research projects on various topics related to linguistics and language. The topics covered are foreign- and native-language instruction, relevant bibliographies, research in contemporary language, communication, textbooks, lexicology, lexicography, linguistics, computational linguistics, machine analysis of language, machine translation, phonetics, psycholinguistics, speech disorders, native language maintenance, language tests, technology and media, and translation science. The project descriptions include the director's name and address and details on project status, goal, collaborators, and funding. (VM)

An open meeting on scientific translation was held to mark the tenth anniversary of the founding of the European Translations Center (ETC) in Delft. The history of ETC and its activities which include collecting, translations and titles, supplying information, publication of "World Index" and "List of Translations", translation services, and cooperative endeavors in the field of terminology were described. A provisional evaluation of an experimental Russian-English Machine Translation (MT) service in cooperation with the ETC and other organizations is provided. (AB)

This report documents results of a two-year effort toward the study and investigation of the design of a prototype system for Chinese-English machine translation in the general area of physics. Previous work in Chinese-English machine translation is reviewed. Grammatical considerations in machine translation are discussed and detailed aspects of the Berkeley grammar II and lexicography are considered. Procedures for interlingual transfer are described in detail. The input and output systems, structure and logic of the syntax analysis system, and auxiliary diagnostic processes are covered as are the means for analyzing processed text. A bibliography is provided along with further details on flow charts, documentation, trees, coding, and sentence generation in the appendixes. (VM)

A Primary Author's Language (PAL-X) has been developed to serve as a documentation language for computer-assisted instructional (CAI) programs. Its development was necessary to permit the dissemination of CAI given the facts that: 1)existing CAI programs were written in over 60 languages; 2)the system for COURSEWRITER II, the most commonly used language, was discontinued by International Business Machines (IBM); and 3)an exchange of programs was found to be nearly impossible. PAL-X was built to contain answer-processing functions, to be independent of hardware and of machine, programing or author languages, to have simple, internationally recognized symbls, and to be capable of partition into subsets. The language was designed to function with non-generative and generative CAI; operations, names and text were separated and the verbosity of programing languages avoided. It was constructed to be used for primary authoring of CAI, to date it has been employed to author CAI material at the University of Freiburg, to translate programs from COURSEWRITER to LIDIA, to introduce students to the principles of CAI, and to document a program in general chemistry. (PB)

Two studies of computer production of mathematical texts and musical scores in Braille analyzed production practices and input preparation problems; the studies also reviewed the Nemeth Code of Braille Mathematics and Scientific Notation and the Revised International Manual of Braille Music. Both studies demonstrated automation to be feasible. Additional studies considered means of advancing computer translation of English Braille, including hyphenation, the 1968 program, proofreading, economic factors, and contraction contexts. A program abstract and a discussion of the potential of the IBM 360 series are included; also appended are a mathematical Braille translation system and a computer program to produce musical Braille. (JD)

Much work in computational linguistics, e.g. the preparation of concordances and text files, has dealt strictly with the surface of language, treating it as nothing more than strings of characters or phonemes. The "classical" scheme, developed as a result of dissatisfaction with the inability of such surface systems to deal with problems such as ambiguity, consists of surface processing, syntactic processing and semantic processing, with the object of obtaining an expression for the content of the input text; work with programming systems for generation of sentences with transformational grammar is representative of this tradition. It must be recognized, however, that the essential characteristic of language is its connection with information and that language is the external manifestation of the human capacity to process symbols in such ways that information is retained. This capacity should be the object of linguistics, and rules of grammar should describe those "action patterns" which underlie human symbol processing. Recent work in applied computational linguistics recognizes the importance of this conception and should therefore lead to wider computer applications, perhaps even to real man-machine conversations and the concomitant use of the computer as an imaginative consultant for a wide range of problems. (FWB)