International Business Machines Corporation

Ahmed M A Nassar

Craig M Trim

Victoria O Odeyemi

Eman Omar

Sarbajit Kumar Rakshit

Barton Wayne Emanuel

Hisham Emad Elshishiny

Albert T Wong

Mohamed Fathy Deyab

Ibrahim M S A Ateya

Tore Notland

Ahmed M M Ragheb Salem

Text is intelligently annotated by first creating a topic map summarizing topics of interest of the user. A data structure is created. The topic map is used to create two linked user dictionaries, a topic dictionary reflecting topic names and a traversal dictionary reflecting the knowledge structure of a topic. Actions may be linked with topic types. When the text to be annotated is being read, the topic data structure of the topics found in the text are automatically instantiated using the dictionaries and any actions previously linked to topic types. Instantiated topic data structures are automatically attached to the text being annotated. A user GUI may be created to allow the user to access and interact with the text annotations.

Intelligent text annotation

For confidence weighting relationships between complex entities in unstructured data an expressed relationship between a subset of a set of tokens is extracted from a knowledge graph corresponding to the unstructured data. The knowledge graph includes the set of tokens. The tokens in the subset of tokens are related in the expressed relationship by a set of predicates. A number of occurrences of the set of predicates in the unstructured data is determined. A number of occurrences of the expressed relationship in the unstructured data is determined. Using the number of occurrences of the set of predicates and the number of occurrences of the expressed relationship, a confidence value is computed and assigned to the expressed relationship.

Confidence weighting of complex relationships in unstructured data

A computer processor determines a root of a first element of a semantic model, in which a first relationship of the first element to a second element of the semantic model, is unknown. The computer processor generates a search token, based on applying morphological rules to the root of the first element and appending a preposition. The computer processor determines one or more regular expressions by applying the search token to search a source of unstructured data. The one or more regular expressions are in a form of a triple, having a subject, a predicate, and an object, and the computer processor applies the predicate of the triple as the first relationship of the first element of the semantic model to a second element of the semantic model.

Augmenting semantic models based on morphological rules

To recombine incorrectly separated tokens in NLP, a determination is made whether a token from an ordered set of tokens is present in a dictionary related to a corpus from which the ordered set is extracted. When the token is not present in the dictionary, and when a compounding threshold has not been reached, the token is agglutinated with a next adjacent token in the ordered set to form the compound token. The compounding threshold limits a number of tokens that can be agglutinated to form a compound token. A determination is made whether the compound token is present in the dictionary. A weight is assigned to the compound token when the compound token is present in the dictionary and a confidence rating of the compound token is computed as a function of the weight. The compound token and the confidence rating are used in NLP of the corpus.

Recombining incorrectly separated tokens in natural language processing

To extract relationships between complex entities from unstructured data, a parser parses, using an existing language model, the unstructured data to generate a parse tree. From the parse tree, a set of tokens is created. A token in the set of tokens includes a set of words found in the unstructured data. The set of tokens is inserted in the existing language model to form an enhanced language model. The unstructured data is re-parsed using the enhanced language model to create a knowledge graph. From the knowledge graph, a relationship between a subset of the set of tokens is extracted.

Growing community of inventors

Katy, TX, United States of America

Ahmed M A Nassar