Context-Aware Prompt and Plug-In Recommendation in AI Conversations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users face difficulty in selecting appropriate plug-ins and prompts in generative artificial intelligence-based conversation services due to the increasing range of available options, leading to inefficient utilization.

Innovation Solution

A method and computing device that predict a user's subsequent query message based on their conversation history and generate recommended prompts or plug-ins using a large language model, conversation prediction model, and recommendation model, enhancing the selection process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the range of plug-ins and prompts is increased to provide extended services, then service functionality is improved, but user difficulty in selecting appropriate plug-ins increases

Engineering Contradiction:
Improveservice functionalityVSAvoidplug-in selection difficulty
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system automatically analyzes user input queries and autonomously recommends appropriate plug-ins and prompts without requiring manual browsing or selection by the user. The recommendation engine processes user queries, matches them with relevant plug-ins based on similarity analysis, and presents tailored recommendations, enabling the system to serve itself in the selection process.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent introduces a recommendation engine as an intermediary component between the user and the plug-in library. This mediator analyzes user intent, compares it with stored plug-in characteristics, and bridges the gap by presenting relevant recommendations, thus simplifying the interaction while maintaining access to diverse plug-ins.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If users manually retrieve and select plug-ins personally, then selection accuracy may be improved, but time consumption and operational complexity increase

Engineering Contradiction:
Improveselection accuracyVSAvoidtime consumption
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of user queries and pre-computes relevant plug-in recommendations before the user needs to make a selection. By anticipating user needs and preparing recommended plug-ins in advance based on query analysis, the system eliminates the time-consuming manual browsing process while maintaining accurate matching.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The recommendation engine incorporates feedback mechanisms where user interactions with recommended plug-ins are analyzed to improve future recommendations. The system learns from user behavior patterns and refines its matching algorithm, thereby improving selection accuracy over time while reducing the effort users need to invest in finding appropriate plug-ins.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4610852A1Method of recommending registered prompt or plug-in in generative artificial intelligence-based conversation service, and computing device using same
Publication Date: 2025.09.03 SAMSUNG SDS CO LTD
  • EP4610852A1 patent drawingFigure 1
  • EP4610852A1 patent drawingFigure 2
  • EP4610852A1 patent drawingFigure 3(a)~3(b)

AI summary

A method of recommending a registered prompt or recommending a plug-in service in a generative artificial intelligence-based conversation service, and a computing device using the same, and the recommendation method may include an operation of generating a predicted query message that is generated by predicting an additional query message to be input by a user, based on a first query message input by the user and a first response message generated in the conversation service in response to the first query message, and an operation of extracting and providing a recommended prompt or a recommended plug-in corresponding to the predicted query message from among a plurality of registered prompts or a plurality of plug-ins.