ftp.cs.arizona.edu/ftp.cs.arizona.edu.tar/ftp.cs.arizona.edu/icon/newsgrp/group93a.txt/000068_icon-group-sender _Thu Feb 18 16:42:03 1993.msg

Internet Message Format | 1993-04-21 | 2KB

Received: by cheltenham.cs.arizona.edu; Thu, 18 Feb 1993 15:25:02 MST
Date: Thu, 18 Feb 93 16:42:03 -0500
From: [e-mail] (Phillip Lee Thomas)
Message-Id: <[e-mail]>
To: [e-mail]
Subject: Text Computing
Status: R
Errors-To: [e-mail]

Re: John Nall's question, What is text processing?

What are the X, Y, and Z's of text processing? Other than saying text
processing is non-numeric computing, which is both most accurate and
very negative, my text processing work has 3 components:

X. Format and edit: these deserve a category of their own because they
take the most time even for numeric computing. Include here the use
of editors as well as use of macros and the programming of editors.

Y. Launder: this is more general than X. Replace, filter, encrypt, code,
and convert data. Coding for part of speech, changing end of line markers
from UNIX to DOS conventions, compressing data, setting data up for
use in a database, verifying (spell checkers included) legal character
sequences are examples.

Z. Retrieve information: locate passages with specific strings or
patterns of strings, find discontinuous information, find information
when strings are incorrectly spelled (fuzzy, soundex searches), build
special purpose databases, find material according to its linguistic
syntax, computational linguistics (ratio of hypotactic versus paratactic
constructions)....

Most computing can be rephrased as text processing if we include logical
operations, e.g., we can do mathematics by directly using lookup tables:
if a pattern of numeric characters is found, replace with values found
in the Add table.

Have fun,
Phillip Lee Thomas ([e-mail]) - [phone]
Library of Congress