Systems and methods for automated user relationship profiling
A computer-implemented method using LLMs and machine learning processes efficiently updates user and family profiles by processing conversation and third-party data, addressing inefficiencies in familial relationship identification and task generation.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- PANASONIC WELL LLC
- Filing Date
- 2025-09-25
- Publication Date
- 2026-05-21
AI Technical Summary
Existing technologies struggle to efficiently and continuously evaluate user data from communications sessions and other sources to define familial relationships among users, leading to inefficiencies in identifying and updating user profiles and family unit dynamics.
A computer-implemented method using large language models and machine learning algorithms to process conversation data, extract key-value pairs of inferred user attributes, and update profiles in real-time, incorporating feedback and third-party data to identify and refine family relationships and generate tasks based on user intents.
Enhances the accuracy and efficiency of user and family profiling by continuously updating profiles with inferred attributes, reducing the need for direct user prompts, and optimizing task recommendations tailored to family goals.
Smart Images
Figure US2025048027_21052026_PF_FP_ABST