Natural Language text Processing Technology
Program name: SYNTPARSE
Natural language parsing is fundamental in all spheres of Text Processing. Without correct parsing of the word, the phrase and the sentence, nothing else in Text Processing works correct. Parsing is needed to analyse the composition of the word and of the sentence. On word level, our Parsers perform accurate morphological analysis of the composition of the word: prefix, root, suffix, grammatical ending and print the Part of Speech of each word, in context: Noun, Verb, Adjective, Conjunction, Adverb, etc. Our Natural Language Parsers recognize also the meaning of the word, in context. On sentence level, our Parsers recognize correctly the composition of the sentence: Subject, Verb and Verbal Tense, Infinitive, Object (direct, indirect), Complement Object, Complement Subject, Verbal Complement, Prepositional Adverbial Clause, Main Clause, Subordinate Clause, Relative Clause, in simple and compound sentences. We use the traditional approach to Natural Language Parsing, taught before Noam Chomsky, because we think, that this approach more accurately describes the composition of the sentence, than the tree-root-like presentation of the composition of the sentence, readily understood by programmers.
A software program of our magnitude has hundreds of rules to resolve the difficulties, mentioned in our books, in order to produce accurate output. There are hundreds, even thousands of exceptions from each of these rules. It is a great challenge to find the exceptions from each rule, in context, and to program these exceptions. That makes each program unique in its performance, compared to similar programs of the competition. In Natural Language Processing, every program is bound, inevitably, to produce, occasionally, wrong analysis of the sentence, to have some errors in the analysis of the text. The programmer discovers these errors during the tests and amends the existing rule or adds a new rule to correct them. Often, the slightest change of a rule in one place, to correct an existing performance error, can lead to unexpected error(s) in another place, where the program has performed, hitherto, correctly. These other errors can be detected only by conducting more tests and checking, each rule, again and again, after each change made in a rule. In other words, a Natural Language Processing task is never final. One can stop and say "so much, so good, for this version!". The accuracy of Natural Language text analysis is the major challenge and the major argument when comparing similar products.
There are 2 versions, older and newer. The older version, for 16 and 32 Bit OS, is with menu and can process files. The newer version, for 32 and 64 Bit OS, is a console application, it can process typed or pasted text input.
