Abstract: A method, computer readable medium and system are provided which retrieve confirming sentences from a sentence database in response to a query. A search engine retrieves confirming sentences from the sentence database in response to the query. IN retrieving the confirming sentences, the search engine defines indexing units based upon the query, with the indexing units including both lemma from the query and extended indexing units associated with the query. The search engine then retrieves a plurality of sentences from the sentence database using the defined indexing units as search parameters. A similarity between each of the plurality of retrieved sentences and the query is determined by the search engine, wherein each similarity is determined as a function of a linguistic weight of a term in the query. The search engine then ranks the plurality of retrieved sentences based upon the determined similarities.

Abstract: A method is disclosed for providing a chunking utility that supports robust natural language processing. A corpus is chunked in accordance with a draft chunking specification. Chunk inconsistencies in the corpus are automatically flagged for resolution, and a chunking utility is provided in which at least some of the flagged inconsistencies are resolved. The chunking utility provides a single, consistent global chunking standard, ensuring compatibility among various applications. The chunking utility is particularly advantageous for non-alphabetic languages, such as Chinese.

Abstract: A system and method identify base noun phrases (baseNP) in a linguistic input. A part-of-speech tagger identifies N-best part-of-speech tag sequences corresponding to the linguistic input. A baseNP identifier identifies baseNPs in the linguistic input using a unified statistical model that identifies the baseNPs, given the N-best POS sequences.

Abstract: A method and apparatus for segmenting text is provided that identifies a sequence of entity types from a sequence of characters and thereby identifies a segmentation for the sequence of characters. Under the invention, the sequence of entity types is identified using probabilistic models that describe the likelihood of a sequence of entities and the likelihood of sequences of characters given particular entities. Under one aspect of the invention, organization name entities are identified from a first sequence of identified entities to form a final sequence of identified entities.

Abstract: The present invention deals with noisy data not by eliminating low frequency dependency structures, but rather by weighting the dependency structures. The dependency structures are weighted to give less weight to dependency structures which are more likely incorrect and to give more weight to dependency structures which are more likely correct.

Abstract: A method is disclosed for providing a chunking utility that supports robust natural language processing. A corpus is chunked in accordance with a draft chunking specification. Chunk inconsistencies in the corpus are automatically flagged for resolution, and a chunking utility is provided in which at least some of the flagged inconsistencies are resolved. The chunking utility provides a single, consistent global chunking standard, ensuring compatibility among various applications. The chunking utility is particularly advantageous for non-alphabetic languages, such as Chinese.

Abstract: Segmentation error candidates are detected using segmentation variations found in an annotated corpus.

Abstract: A method, computer readable medium and system are provided which retrieve confirming sentences from a sentence database in response to a query. A search engine retrieves confirming sentences from the sentence database in response to the query. IN retrieving the confirming sentences, the search engine defines indexing units based upon the query, with the indexing units including both lemma from the query and extended indexing units associated with the query. The search engine then retrieves a plurality of sentences from the sentence database using the defined indexing units as search parameters. A similarity between each of the plurality of retrieved sentences and the query is determined by the search engine, wherein each similarity is determined as a function of a linguistic weight of a term in the query. The search engine then ranks the plurality of retrieved sentences based upon the determined similarities.

Abstract: A method, computer readable medium and system are provided which retrieve confirming sentences from a sentence database in response to a query. A search engine retrieves confirming sentences from the sentence database in response to the query. IN retrieving the confirming sentences, the search engine defines indexing units based upon the query, with the indexing units including both lemma from the query and extended indexing units associated with the query. The search engine then retrieves a plurality of sentences from the sentence database using the defined indexing units as search parameters. A similarity between each of the plurality of retrieved sentences and the query is determined by the search engine, wherein each similarity is determined as a function of a linguistic weight of a term in the query. The search engine then ranks the plurality of retrieved sentences based upon the determined similarities.

Abstract: A method for optimizing a language model is presented comprising developing an initial language model from a lexicon and segmentation derived from a received corpus using a maximum match technique, and iteratively refining the initial language model by dynamically updating the lexicon and re-segmenting the corpus according to statistical principles until a threshold of predictive capability is achieved.

Abstract: A system and method identify base noun phrases (baseNP) in a linguistic input. A part-of-speech tagger identifies N-best part-of-speech tag sequences corresponding to the linguistic input. A baseNP identifier identifies baseNPs in the linguistic input using a unified statistical model that identifies the baseNPs, given the N-best POS sequences.

Abstract: The present invention relates to a corpus for use in training a language model. The corpus includes a plurality of characters and a plurality of morphological tags associated with a plurality of sequences of characters. The plurality of morphological tags indicate a morphological type of an associated sequence of characters and a combination of parts forming a morphological subtype.

Abstract: A system and method identify base noun phrases (baseNP) in a linguistic input. A part-of-speech tagger identifies N-best part-of-speech tag sequences corresponding to the linguistic input. A baseNP identifier identifies baseNPs in the linguistic input using a unified statistical model that identifies the baseNPs, given the N-best POS sequences.

Abstract: The present invention deals with noisy data not by eliminating low frequency dependency structures, but rather by weighting the dependency structures. The dependency structures are weighted to give less weight to dependency structures which are more likely incorrect and to give more weight to dependency structures which are more likely correct.

Abstract: A method and apparatus for segmenting text is provided that identifies a sequence of entity types from a sequence of characters and thereby identifies a segmentation for the sequence of characters. Under the invention, the sequence of entity types is identified using probabilistic models that describe the likelihood of a sequence of entities and the likelihood of sequences of characters given particular entities. Under one aspect of the invention, organization name entities are identified from a first sequence of identified entities to form a final sequence of identified entities.

Abstract: A method for optimizing a language model is presented comprising developing an initial language model from a lexicon and segmentation derived from a received corpus using a maximum match technique, and iteratively refining the initial language model by dynamically updating the lexicon and re-segmenting the corpus according to statistical principles until a threshold of predictive capability is achieved.

Abstract: A method, computer readable medium and system are provided which retrieve confirming sentences from a sentence database in response to a query. A search engine retrieves confirming sentences from the sentence database in response to the query. IN retrieving the confirming sentences, the search engine defines indexing units based upon the query, with the indexing units including both lemma from the query and extended indexing units associated with the query. The search engine then retrieves a plurality of sentences from the sentence database using the defined indexing units as search parameters. A similarity between each of the plurality of retrieved sentences and the query is determined by the search engine, wherein each similarity is determined as a function of a linguistic weight of a term in the query. The search engine then ranks the plurality of retrieved sentences based upon the determined similarities.

Abstract: The present invention performs machine translation by matching fragments of a source language sentence to be translated to source language portions of an example in example base. When all relevant examples have been identified in the example base, the examples are subjected to phrase alignment in which fragments of the target language sentence in each example are aligned against the matched fragments of the source language sentence in the same example. A translation component then substitutes the aligned target language phrases from the matched examples for the matched fragments in the source language sentence.

Abstract: A system and method identify base noun phrases (baseNP) in a linguistic input. A part-of-speech tagger identifies N-best part-of-speech tag sequences corresponding to the linguistic input. A baseNP identifier identifies baseNPs in the linguistic input using a unified statistical model that identifies the baseNPs, given the N-best POS sequences.