US2018292221A1PendingUtilityA1

Deep learning allergen mapping

Assignee: IBMPriority: Apr 5, 2017Filed: Sep 14, 2017Published: Oct 11, 2018
Est. expiryApr 5, 2037(~10.7 yrs left)· nominal 20-yr term from priority
G01C 21/3461G06F 17/30864G01C 21/3691G01C 21/3415G06N 99/005G06N 3/0464G06N 3/0895G06N 3/09G16H 70/60G16H 20/60G06N 5/022G16H 10/60G06F 16/951G06F 16/29G06N 20/00G06N 3/04
51
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

An entry on an allergen map may be generated by a computer system where a deep learning model is trained using online content data. Allergen content data which contains geographic data may be detected from the online content data. The allergen content data may be analyzed by the computer system and tagged with a quality and intensity indicator. Based on the tagging and the geographic location, an allergen map may be generated.

Claims

exact text as granted — not AI-modified
1 . A method for generating an allergen recommendation for a user, the method comprising:
 providing, to a deep learning model, wherein the deep learning model is an artificial neural network, comprising a set of algorithms carried out over a set of one or more processors of a computer system, a set of known online content data, wherein the deep learning model is trained using the known online content data, and wherein the known online content data comprises a set of known allergen data, and wherein the deep learning model is trained by applying object recognition and natural language processing to the allergen content data, and wherein online content data:
 is collected via data mining across the Internet, and 
 comprises photographic content and video content; 
   receiving, from a user device, a set of unknown online content data;   detecting, by the computer system, from the set of unknown online content data, allergen content data, wherein the allergen content data comprises a geographic location, and wherein the online content data is a particular allergen source comprising a particular type of plant, and wherein the allergen content data is detected from a background of the photographic content;   analyzing, by the deep learning model of the computer system, the allergen content data;   tagging, by the computer system, based on the analyzing, the allergen content data, wherein the tagging comprises a quality indicator and an intensity indicator, wherein the quality indicator conveys a type of allergen detected in the allergen content data and the intensity indicator conveys a particular level of intensity of an allergen detected in the allergen content data;   generating, by the computer system and based on the tagging and the geographic location, an entry on an allergen map;   generating, by the computer system and in response to the generating the entry on the allergen map, a set of new allergen entries on the allergen map;   generating, by the computer system and based on the allergen map, a recommendation for a user, wherein the recommendation is responsive to a user profile, wherein the user profile comprises data specific to the user including specific allergen sensitivities of the user, a location of the user, and a travel route of a user, and wherein the recommendation is “stay indoors”; and   transmitting, to the user device, the recommendation.

Join the waitlist — get patent alerts

Track US2018292221A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.