# US 10,482,393 B2 · Machine-based learning systems, methods, and apparatus for interactively mapping raw data objects to recognized data objects > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,482,393 B2 - Document ID: US10482393B2 - Status: Granted - Kind: B2 - Application number: 15/591,610 - Filed: 2017-05-10 - Published: 2018-11-15 - Granted: 2019-11-19 - Earliest claimed: 2017-05-10 - Assignees: The Travelers Indemnity Company - Applicants: The Travelers Indemnity Company - Inventors: Alex A. Shyshkov; Sarasaleka Devasenan - Web page: https://patentgrep.com/patent/US10482393B2 ## CPC - G06N 20/00: Machine learning - G06F 16/583: using metadata automatically derived from the content - G06F 18/24133: {Distances to prototypes} - G06V 30/19173: {Classification techniques} - G06V 30/413: Classification of content, e.g. text, photographs or tables - G06F 40/205: Parsing - G06V 2201/10: Recognition assisted with metadata - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Systems, apparatus, methods, and articles of manufacture provide for analysis of raw data and conversion of the raw data to searchable data objects. In one embodiment, a formatted searchable data object is generated in a first interface portion of a user interface in conjunction with the corresponding raw data displayed in a second interface portion of the user interface. ## Claims ### Claim 1 A machine-based learning system comprising: a computerized mobile device comprising: a processor, a touchscreen input device in communication with the processor, a microphone in communication with the processor, an image capture device in communication with the processor, a wireless communications interface in communication with the processor, and a computer-readable memory in communication with the processor, the computer-readable memory storing a raw data capture application that when executed by the processor directs the processor to: receive raw data via at least one of the touchscreen input device, the microphone, and the image capture device; and transmit, using the wireless communications interface, the raw data to a data conversion and learning server; and the data conversion and learning server, comprising: a second processor; a second computer-readable memory in communication with the second processor, the second computer-readable memory storing instructions that when executed by the second processor direct the second processor to: receive, from the computerized mobile device, the raw data; convert the raw data into at least one searchable data object using at least one data parser; access a database of a plurality of predefined data query objects; select a searchable data object of the at least one searchable data object converted from the raw data; identify at least one predefined data query object of the plurality of predefined data query objects, based on the selected searchable data object; format the selected searchable data object based on the identified at least one predefined data query object; and store at least a respective portion of the selected searchable data object in association with each identified at least one predefined data query object, wherein formatting the selected searchable data object based on the identified at least one predefined data query object comprises: generating a first interface portion of a user interface including the formatted searchable data object; and generating a second interface portion including the raw data from which the formatted searchable data object was converted. ### Claim 2 (depends on 1) The system of claim 1, wherein converting the raw data into at least one searchable data object comprises: selecting a text description for at least one physical object indicated in the raw data. ### Claim 3 (depends on 1) The system of claim 1, wherein converting the raw data into at least one searchable data object comprises: identifying at least one physical object displayed in a video file of the raw data by comparing the video file to stored object image data using a video analysis parser; and selecting a text description for the at least one identified physical object. ### Claim 4 (depends on 1) The system of claim 1, wherein converting the raw data into at least one searchable data object comprises: identifying at least one physical object displayed in an image file of the raw data by comparing the image file to stored object image data using an image analysis parser; and selecting a text description for the at least one identified physical object. ### Claim 5 (depends on 1) The system of claim 1, wherein converting the raw data into at least one searchable data object comprises: identifying at least one text description in the raw data using at least one of the following: an electronic handwriting parser, and an optical character recognition parser. ### Claim 6 (depends on 1) The system of claim 1, wherein converting the raw data into at least one searchable data object comprises: identifying at least one audio description in the raw data using a voice recognition parser; and selecting a text description for the at least one audio description. ### Claim 7 (depends on 1) The system of claim 1, wherein converting the raw data into at least one searchable data object comprises: identifying at least one barcode in the raw data using a barcode parser; and selecting barcode information associated with the at least one barcode. ### Claim 8 (depends on 1) The system of claim 1, wherein the identified at least one predefined data query object comprises a predefined form field for an information intake process. ### Claim 9 (depends on 1) The system of claim 1, wherein identifying the at least one predefined data query object based on the selected searchable data object comprises: comparing a first text description associated with the selected searchable data object to a respective second text description associated with each predefined data query object. ### Claim 10 A machine-based learning method, the method comprising: receiving, using a raw data capture application, raw data via at least one of a touchscreen input device, a microphone, and an image capture device of a computerized mobile device; transmitting, using a wireless communications interface of the computerized mobile device, the raw data to a data conversion and learning server; receiving, by the data conversion and learning server from the computerized mobile device, the raw data; converting, by the data conversion and learning server, the raw data into at least one searchable data object using at least one data parser; accessing, by the data conversion and learning server, a database of a plurality of predefined data query objects; selecting, by the data conversion and learning server, a searchable data object of the at least one searchable data object converted from the raw data; identifying, by the data conversion and learning server, at least one predefined data query object of the plurality of predefined data query objects, based on the selected searchable data object; formatting, by the data conversion and learning server, the selected searchable data object based on the identified at least one predefined data query object; and storing, by the data conversion and learning server, at least a respective portion of the selected searchable data object in association with each identified at least one predefined data query object, wherein formatting the selected searchable data object based on the identified at least one predefined data query object comprises: generating a first interface portion of a user interface including the formatted searchable data object; and generating a second interface portion including the raw data from which the formatted searchable data object was converted. ### Claim 11 (depends on 10) The method of claim 10, wherein converting the raw data into at least one searchable data object comprises: selecting a text description for at least one physical object indicated in the raw data. ### Claim 12 (depends on 10) The method of claim 10, wherein converting the raw data into at least one searchable data object comprises: identifying at least one physical object displayed in a video file of the raw data by comparing the video file to stored object image data using a video analysis parser; and selecting a text description for the at least one identified physical object. ### Claim 13 (depends on 10) The method of claim 10, wherein converting the raw data into at least one searchable data object comprises: identifying at least one physical object displayed in an image file of the raw data by comparing the image file to stored object image data using an image analysis parser; and selecting a text description for the at least one identified physical object. ### Claim 14 (depends on 10) The method of claim 10, wherein converting the raw data into at least one searchable data object comprises: identifying at least one text description in the raw data using at least one of the following: an electronic handwriting parser, and an optical character recognition parser. ### Claim 15 (depends on 10) The method of claim 10, wherein converting the raw data into at least one searchable data object comprises: identifying at least one audio description in the raw data using a voice recognition parser; and selecting a text description for the at least one audio description. ### Claim 16 (depends on 10) The method of claim 10, wherein converting the raw data into at least one searchable data object comprises: identifying at least one barcode in the raw data using a barcode parser; and selecting barcode information associated with the at least one barcode. ### Claim 17 (depends on 10) The method of claim 10, wherein the identified at least one predefined data query object comprises a predefined form field for an information intake process. ### Claim 18 (depends on 10) The method of claim 10, wherein identifying the at least one predefined data query object based on the selected searchable data object comprises: comparing a first text description associated with the selected searchable data object to a respective second text description associated with each predefined data query object.