Back to results
US 11,893,045 B1Granted
Systems and methods for pattern-based multi-stage deterministic data classification
AssigneeThe Travelers Indemnity Company
InventorZi Cheng Feng
Life of this patent
Estimating the term…
Abstract
Systems and methods for pattern-based multi-stage deterministic data classification that may reduce processing and memory overhead while providing more accurate data classifications.
Claims20
A system for pattern-based multi-stage deterministic data classification, comprising:
- a server comprising a plurality of electronic processing devices;
- and
- a non-transitory data storage device in communication with the server, the non-transitory data storage device storing (i) an Optical Character Recognition (OCR) algorithm, (ii) a listing of keywords representing required data elements for a consolidated set of strategic classification rules, (iii) a mapping of a plurality of data availability categories to a plurality of signature analysis routines, and (iv) pattern-based multi-stage deterministic data classification instructions that when executed by the electronic processing devices, result in:
- receiving an application for an underwriting product, the application comprising at least one data value and at least one attached file;
- converting, utilizing the OCR algorithm, at least one portion of the at least one attached file into at least one searchable text character, thereby defining a searchable version of the at least one attached file;
- searching, utilizing the listing of keywords representing the required data elements for the consolidated set of strategic classification rules, the searchable version of the at least one attached file;
- identifying, based on the searching, and for each required data element for the consolidated set of strategic classification rules, whether a value for the required data element exists in the at least one attached file;
- categorizing, based on the identifying of whether the values for the required data elements exist in the at least one attached file, the application into one of a plurality of pre-defined data availability categories;
- selecting, based on the categorized one of the plurality of pre-defined data availability categories for the application and utilizing the mapping of the plurality of data availability categories to the plurality of signature analysis routines, one of the plurality of signature analysis routines;
- identifying, by an execution of the selected one of the plurality of signature analysis routines, values for the required data elements for the consolidated set of strategic classification rules;
- evaluating, based on the identified values for the required data elements for the consolidated set of strategic classification rules, the consolidated set of strategic classification rules;
- identifying, based on the evaluating, a numeric value for a classification;
- and
- automatically routing the application for the underwriting product to an address assigned to the identified classification.
11 more elements
A system for pattern-based multi-stage deterministic data classification, comprising:
- a server comprising a plurality of electronic processing devices;
- and
- a non-transitory data storage device in communication with the server, the non-transitory data storage device storing (i) an Optical Character Recognition (OCR) algorithm, (ii) a listing of keywords representing required data elements for a consolidated set of strategic classification rules, wherein the consolidated set of strategic classification rules are defined utilizing a pattern-based multi-stage deterministic consolidation process, (iii) a mapping of a plurality of data availability categories to a plurality of signature analysis routines, and (iv) pattern-based multi-stage deterministic data classification instructions that when executed by the electronic processing devices, result in:
- receiving an application for an underwriting product, the application comprising at least one data value and at least one attached file;
- converting, utilizing the OCR algorithm, at least one portion of the at least one attached file into at least one searchable text character, thereby defining a searchable version of the at least one attached file;
- searching, utilizing the listing of keywords representing the required data elements for the consolidated set of strategic classification rules, the searchable version of the at least one attached file;
- identifying, based on the searching, and for each required data element for the consolidated set of strategic classification rules, whether a value for the required data element exists in the at least one attached file;
- categorizing, based on the identifying of whether the values for the required data elements exist in the at least one attached file and based on the at least one data value of the application, the application into one of a plurality of pre-defined data availability categories;
- selecting, based on the categorized one of the plurality of pre-defined data availability categories for the application and utilizing the mapping of the plurality of data availability categories to the plurality of signature analysis routines, one of the plurality of signature analysis routines;
- identifying, by an execution of the selected one of the plurality of signature analysis routines, values for the required data elements for the consolidated set of strategic classification rules;
- evaluating, based on the identified values for the required data elements for the consolidated set of strategic classification rules, the consolidated set of strategic classification rules;
- identifying, based on the evaluating, a numeric value for a classification;
- and
- automatically routing the application for the underwriting product to an address assigned to the identified classification.
11 more elements
Dependencies are inferred from the claim text.