Hartford Fire Insurance Compnay

Vladislav Michael Beznos

Arthur Paul Drennan, Iii

David J Kuethman

William C Lewis

Fred A Robinson

Lynne C Clark

Tracey Ellen Steger

Steven J Cultrera

Saira A Kazmi

Brenda K Boles

A device and method are described for a universal analytical data mart and data structure for same. The analytical data mart (ADM) associated data structure is designed to allow data from disparate sources to be integrated, enabling streamlined business intelligence, reporting and ad hoc analysis. Conceptually, the ADM enables analytics and business intelligence from multiple frames of reference including people, such as parties and actors including individuals and organizations, places, such as addresses with geographic information at various levels of view, objects, such as insured properties, automobiles and machinery, and events, milestones which happen at points in time and provide analytical/business value.

Universal analytical data mart and data structure for same

In some embodiments, image input data is received from multiple sources. The received image input data may then be aggregated and mapped to create a set of image input data. An event in the set of image input data may be automatically detected, such as by being triggered by a rule and an associated tag. An image mining result database may be updated by adding an entry to the database identifying each detected event and the triggering rule. An indication associated with the image mining result database may then be transmitted to a plurality of risk applications.

System and method for evaluating images to support multiple risk applications

A device and method are described for universally repositing repeatedly accessible information. The system includes a plurality of different data sources providing data stored in a plurality of storage devices, at least a first processor for searching each of the plurality of data sources using a plurality of data mining techniques and identifying selected portions of each data source, wherein the first processor further identifies at least one pattern in at least one of the selected portions, wherein the data mining techniques comprise at least machine learning and text mining, a database for staging the selected portions and patterns, wherein the staging includes preparation, reorganization an management of the selected portions and patterns, and at least a second processor for assembling the staged materials by incorporating analytics within the staged data, the second processor executing instructions stored in a storage medium to run computer models on the assembled staged data to provide information about the assembled staged data, and based on defined criteria produce a ranking added the assembled staged data and store the assembled staged data with the associated ranking to universally reposit repeated access to the information and automatically activating at least two systems by providing the associated ranking directly via at least one interface to the at least two systems to allow access to the stored information with the associated ranking.

Universal repository for holding repeatedly accessible information

Growing community of inventors

Longmeadow, MA, United States of America

Vladislav Michael Beznos