First American Financial Corporation

Calvin Powell

Raza Khan

Chris Rubio

Arun Rangarajan

Madhu Kolli

Zheqi Tan

Prabhaker Narsina

Some implementations of the disclosure describe a non-transitory computer-readable medium having executable instructions stored thereon that, when executed by a processor, cause the processor to perform operations comprising: obtaining a document image file including a first image corresponding to a first page of a document; generating, using a first trained model, based on the first image, a first prediction that the first page includes a term identifying a class of people, the first prediction including a first location of the term within the first image; in response to generating the first prediction that the first page includes the term identifying the class of people, generating, using a second trained model, based on the first image, a second prediction of whether or not the first page includes a section that uses a term identifying a class of people in a specific context.

AI-based detection of contextual class description in document images

Some implementations of the disclosure are directed to: extracting metadata from textual data representations of a plurality of document images, and contextualizing the extracted metadata; storing the extracted metadata and the textual data representations in a full text index database; and transferring the extracted metadata and the textual data representations from the full text index database to a search engine platform, the search engine platform indexing and storing the transferred extracted metadata to allow for searching of the indexed, extracted metadata, the indexed, extracted metadata having been correlated to the textual data representations, where the search engine platform allows for the selection of extracted metadata stored in full text index database that is transferred to the search engine platform.

System and method for data extraction and searching

Systems and methods are provided for quickly and efficiently searching and receiving results for real estate-related information without or at least with minimal human processing of real estate-related documents. Optical character recognition on a plurality of scanned document images is performed to obtain a plurality of textual data representations of the real estate-related documents. Data is extracted from the textual data representations, and subsequently contextualized according to a real estate-related context. Aspects of the extracted data as well as the textual data representations are provided as search results based on one or more searches for real estate-related information.

Growing community of inventors

Villa Park, CA, United States of America

Calvin Powell