Chatbot Diet Assistance via Intent Deduction and Knowledge Graph

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users face challenges in obtaining personalized and efficient diet-related information, such as food recommendations and recipes, as existing methods like browsing health websites or search engines provide general information that is not tailored to individual health conditions or cooking preferences, and are time-consuming.

Innovation Solution

An AI chatbot system that deduces user diet intentions and requirements from session messages, extracts relevant information, and provides personalized diet suggestions, including recipe recommendations and product information, using a knowledge graph and user profile data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If users browse health websites or use search engines to obtain diet information, then they can access general diet information, but the information is not personalized and the process is time-consuming

Engineering Contradiction:

Improvepersonalization of diet informationVSAvoidtime required to search for diet information

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-collecting user health data, cooking preferences, and dietary restrictions during registration. It pre-processes this information to create a personalized user profile that enables rapid generation of tailored diet suggestions without requiring users to search repeatedly

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The chatbot enables self-service by automatically analyzing user inputs, deducing diet intentions, and generating personalized recommendations without requiring users to manually search multiple websites or compile information themselves

Inventive Principle:

Principle #25Self-service

2Productivity

If a chatbot provides general responses based on keyword matching, then it can respond quickly, but the responses lack personalization and relevance to user health conditions

Engineering Contradiction:

Improveresponse speed of chatbotVSAvoidpersonalization of diet suggestions

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary action by pre-collecting user health data, cooking preferences, and dietary restrictions during registration. It pre-processes this information to create a personalized user profile that enables rapid generation of tailored diet suggestions without requiring users to search repeatedly

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms by continuously learning from user interactions, dietary outcomes, and preference adjustments. This allows the chatbot to refine its personalization accuracy over time while maintaining efficient response generation through optimized algorithms

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11430347B2Providing diet assistance in a session

Publication Date: 2022.08.30 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

The present disclosure provides method and apparatus for providing diet assistance to a user in a session. At least one message may be received in the session, the session being between the user and an electronic conversational agent. A diet intention associated with the user may be deduced based on the session and the at least one message. Diet information may be extracted from the at least one message. A diet requirement of the user may be identified based at least on the session and the at least one message. A diet suggestion for the user may be obtained based at least on the diet intention, the diet information and the diet requirement. A response including the diet suggestion may be generated and provided to the user.