Direct navigation for information retrieval
First Claim
Patent Images
1. A computer-implemented method of retrieving information, comprising:
- performing a pre-processing stage by parsing documents contained in a collection with a grammar in order to identify one or more concepts contained therein, and assigning concept labels to the documents contained in the collection based on the identified concepts; and
performing a post-processing stage by applying the grammar to a query to convert the query to one or more concepts and mapping the concepts to the concept labels that match the concepts, wherein the query is normalized, the normalized query is parsed and converted into fragments according to a feature lexicon, the fragments are inflated by selectively merging state information provided by a session service with a meaning representation for the query, and the inflated fragments are converted into a meaning resolution through a meaning resolution stage that determines whether there is a valid interpretation of a key-value grouping of each of the fragments, such that the meaning resolved fragments are associated with the concepts.
6 Assignments
0 Petitions
Accused Products
Abstract
A method of document retrieval is provided. The method includes assigning concept labels to documents contained in a collection according to grammar rules, receiving a query, converting the query to a query concept using the grammar rules, and mapping the query concept to a concept label.
141 Citations
15 Claims
-
1. A computer-implemented method of retrieving information, comprising:
-
performing a pre-processing stage by parsing documents contained in a collection with a grammar in order to identify one or more concepts contained therein, and assigning concept labels to the documents contained in the collection based on the identified concepts; and performing a post-processing stage by applying the grammar to a query to convert the query to one or more concepts and mapping the concepts to the concept labels that match the concepts, wherein the query is normalized, the normalized query is parsed and converted into fragments according to a feature lexicon, the fragments are inflated by selectively merging state information provided by a session service with a meaning representation for the query, and the inflated fragments are converted into a meaning resolution through a meaning resolution stage that determines whether there is a valid interpretation of a key-value grouping of each of the fragments, such that the meaning resolved fragments are associated with the concepts. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11)
-
-
12. A computer-implemented method of document retrieval, comprising:
-
performing a pre-processing stage by parsing documents contained in a collection according to grammar rules in order to identify one or mote concepts contained therein, and assigning concept labels to the documents contained in the collection according to the grammar rules based on the identified concepts; and performing a post-processing stage by applying the grammar rules to a query to convert the query to one or more concepts and mapping the concepts to the concept labels that match the concepts, wherein the query is normalized, the normalized query is parsed and converted into fragments according to a feature lexicon, the fragments are inflated by selectively merging state information provided by a session service with a meaning representation for the query, and the inflated fragments are converted into a meaning resolution through a meaning resolution stage that determines whether there is a valid interpretation of a key-value grouping of each of the fragments, such that the meaning resolved fragments are associated with the concepts. - View Dependent Claims (13, 14, 15)
-
Specification