Mitsubishi Electric Research Laboratories, Inc.

Mitsubishi Electric Information Technology Center America, Inc.

Emmanuel Roche

Yves Schabes

Andrew R Golding

A context-based dictionary entry access system includes selecting and rang an entry based on the part-of-speech assigned to the given word in a sentence. The entries corresponding to the word in context are first selected, followed by availability of entries that are not relevant to the current context. The part of speech of the given word in context is selected based on the utilization of a part-of-speech tagger such that only those entries are selected in which the corresponding word makes syntactic sense in the sentence.

Context based system for accessing dictionary entries

In a grammar checking system in which a sentence is first tagged as to pa of speech, the probability of the sequence of the parts of speech being correct is utilized to correct improper use of troublesome words, especially those identical sounding words which are spelled differently. The system corrects word usage based not on the probability of the entire sentence being correct but rather on the probability of the sequence of the parts of speech being correct. As part of the subject invention, the parts of speech sequence probability is utilized in part of speech sequence verification, underlying spelling recovery, auxiliary verb correction, determiner correction, and in a context-sensitive dictionary lookup.

System for correcting grammar based on parts of speech probability

A system for assigning part-of-speech tags to English text includes an improved contextual tagger which utilizes a deterministic finite state transducer to improve tagging speed such that large documents can have its sentences accurately tagged as to parts of speech to permit fast grammar checking, spell checking, information retrieval, text indexing and optical character recognition. The subject system performs by first acquiring a set of rules by examining a training corpus of tagged text. Then, these rules are transformed into a deterministic finite-state transducer through the utilization of non-deterministic transducers, a composer and a determiniser. In order to tag an input sentence, the sentence is initially tagged by first assigning each word in the sentence with its most likely part of speech tag regardless of the surrounding words in the sentences. The deterministic finite-state transducer is then applied on the resulting sequence of part of speech tags using the surrounding words and obtains the final part of speech tags. The Subject System requires an amount of time to compute the part-of-speech tags which is proportional to the number of words in the input sentence and which is independent of the number of rules it has applied.

Contextual tagger utilizing deterministic finite state transducer

In a grammar checking system in which a sentence is first tagged as to parts of speech, the probability of the sequence of the parts of speech being correct is utilized to correct improper use of troublesome words, especially those identical sounding words which are spelled differently. The system corrects word usage based not on the probability of the entire sentence being correct but rather on the probability of the sequence of the parts of speech being correct. As part of the subject invention, the parts of speech sequence probability is utilized in part of speech sequence verification, underlying spelling recovery, auxiliary verb correction, determiner correction, and in a context-sensitive dictionary lookup.

System for correcting grammer based parts on speech probability

In a grammar checking system which includes first tagging a sentence as to parts of speech, a module is provided for correcting incorrect auxiliary verb sequences involving two or more verbs by first detecting the starting and ending points of the error and then correcting the indicated verbs through the utilization of a directed acyclic graph containing the set of correct auxiliary verb sequences constructed for all possible such sequences. Thereafter, a further directed graph functions as a finite state transducer for proposing corrections for incorrect auxiliary verb sequences as determined by the aforementioned acyclic graph of incorrect verb sequences.

System for correcting auxiliary verb sequences

In a grammar checking system which includes first tagging a sentence as to parts of speech and recovery of underlying spellings, a system for the correction of inflections includes identifying typical example of words which do not follow normal rules either as to pluralization, past tense, past participle, comparative formation and superlative formation, with this unique list of incorrect words forming the basis for suggesting suitable replacement words the system first performs a conventional spell check to detect mispellings by dictionary lookup system. Thereafter correct words are suggested based on an English dictionary and on the root forms and the morphological features derived from the incorrect words in the compendium.

Word inflection correction system

In a grammar checking system which includes first tagging a sentence as to parts of speech, underlying spelling is recovered by removing the effects of capitalization of a word so that appropriate inflection and or spelling can be suggested by the system. In order to determine the underlying spelling the system determines whether a noun is a proper noun through the utilization of a part of speech tagger and the utilization of part of speech trigram probabilities, with capitalized and uncapitalized versions of the word having different trigram probabilities. The system also establishes whether a word is an ordinary word as opposed to a proper noun or other intrinsically capitalized word. With the system further determining which of two interpretations of the word is the best one.

System for underlying spelling recovery

In a grammar checking system which includes first tagging a sentence as to parts of speech, an improper determiner correction module detects improper referents of a noun phrase and suggests insertion of a determiner should one be necessary or the deletion of an improper determiner. To detect improper use of a determiner, parts of speech tags are utilized to characterize a sentence, thus to identify noun phrases by maximally matching a pattern that defines which sequences of parts of speech tags constitute valid noun phrases. This is accomplished by identifying the start of the noun phrase and its end to permit checking for either missing determiners, extraneous determiners, or lack of agreement in number for the constituents of the noun phrase. Once a noun phrase is found, the system tests to see if the noun is a head noun and thereafter if the head noun is a singular non-proper noun, whether the noun phrase contains a determiner, whether the noun phrase is a title, whether the head noun is a mass noun or whether the noun phrase is part of an idiom. The system also checks to see if the head noun is a proper noun and if so whether the noun phrase contains a determiner. Finally, if the noun phrase contains a determiner the system checks if the determiner agrees in number with the head noun. In this manner, both missing determiners, extraneous determiners and determiner noun number disagreement are detected and reported.

Growing community of inventors

Boston, MA, United States of America

Emmanuel Roche