PLNLP(ID:5713/pln001)

for Portable Language for Natural Language Processing


Pronounced "Penelope"!


Programmign language effort from IBM Yorktown Hieghts to develop an integrated system for NLP

The research team that led he effort emigrated to Redmond and continue the work on PEG, Critique etc there
References:
  • Brink, Dan "MLJ Computer Corner" The Modern Language Journal 70(2) Summer, 1986 pp149-152 view details Extract: High Level Languages
    High Level Languages
    High level languages are characterized by a format and style which allow programmers to work within a framework which they are familiar with, one which allows them to specify what they Want done in the Same way they would if they were doing the problem with pencil and Paper. The idea is to keep the computer from intruding as much as possible. Some of the popular early languages of this type include FORTRAN and COBOL, and, for that matter, BASIC; more recent systems include C, SNOBOL4, and PROLOG. And many more such systems exist.
    If this is the category which seems the most appealing to you, then you must decide which one to learn. The clear choice of a high level language oriented toward language analysis has not yet established itself, but numerous possibilities have begun to appear. They tend to fall into two general categories: 1) those associated with the branch of artificial intelligence research concerned with the computer processing of natural language processing (NLP); 2) those associated with computer-aided language learning (CALL). Several popular languages exist in the former group: 1) LISP (especially the dialect INTERLISP, used in the XEROX AI workstation); 2) PROLOG, used by a wide variety of computational linguists, especially in Europe and Japan; 3) PLNLP, recently developed by IBM specifically as a programming language for natural language processing. The most ambitious work in the development of a computer language for instruction is probably being done in the CALICO project at Brigham Young University, but numerous other groups are working in the area as well, especially in England.
  • Jensen, Karen; Heidorn, George E. and Richardson, Stephen D. (eds) "Natural Language Processing: The PLNLP Approach" Kluwer Academic Publishers, Boston, 1993. view details Abstract: Natural Language Processing: The PLNLP Approach" describes a decade of computational linguistic work at IBM Research. A very broad-coverage NLP system, including a programming language (PLNLP), development tools, and analysis and synthesis components, was developed and incorporated into a variety of well-known practical applications ranging from text critiquing (CRITIQUE) to machine translation (e.g., SHALT). This book is the first published collection of papers describing this system and its use. Twenty-six authors from nine countries contributed to this volume.
  • Jacobs, Paul S. review of Jensen et al 1993 in Computational Linguistics Volume 19, Number 4 1994 view details Extract: Details of the PLNLP project
    PLNLP (often pronounced "Penelope" or "Plenelope," the editors point out) is the
    basis for some sizable applied natural language projects, the best known of which is
    the Critique text-processing system, a grammar and style checker that IBM took into
    development. Within PLNLP, the IBM group developed an extensive English grammar
    called PEG (PLNLP English Grammar) and a text-critiquing system called EPISTLE,
    which later became Critique. By the time Critique (along with the book's editors)
    moved from Yorktown to an IBM development lab in Bethesda, a variety of simpler,
    relatively inexpensive text-critiquing competitors had come to market through small
    companies. The editors now work for the Research Division of Microsoft.
    Resources