Patent · US Active

Intelligent recommendations implemented by modelling user profile through deep learning of multimodal user data

US11164105B2 · kind B2 · utility

0Cited by
13References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateNov 13, 2017
Grant dateNov 2, 2021
Priority date
Expiry dateSep 3, 2040

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N5/022
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Systems and methods are provided to implement intelligent recommendations to users by modeling user profiles through deep learning of multimodal user data. For example, a recommendation computing platform collects multimodal user data from a computing device of a registered user, wherein the multimodal user data include time-series data, unstructured textual data, and multimedia data. A first deep learning classification engine is utilized to extract features from the multimodal user data. A second deep learning classification engine is utilized to generate a profile of the registered user based on the extracted features. A deep recommendation classification engine is utilized to determine a recommendation for the registered user based on the profile of the registered user, wherein the recommendation identifies at least one additional registered user. The recommendation is presented to the registered user on the computing device of the registered user.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.