Method of recommending registered prompt or plug-in in generative artificial intelligence-based conversation service, and computing device using same
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulty in selecting appropriate plug-ins and prompts in generative artificial intelligence-based conversation services due to the increased range of use and number of available plug-ins, leading to inefficient personal retrieval and selection processes.
Innovation Solution
A method and computing device that generate a predicted query message based on user input and response, using a large language model, conversation history database, and machine learning models to recommend relevant prompts or plug-ins, optimizing the selection process through LLM-based and similarity-based predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If users personally retrieve and select plug-ins, then users can select plug-ins according to their own preferences, but the selection process becomes time-consuming and inefficient as the number of plug-ins increases
Solution Approach 1:
The system automatically performs the retrieval and selection of plug-ins and prompts based on conversation context, eliminating the need for users to manually search through numerous options. The AI analyzes the conversation history and automatically identifies relevant plug-ins and prompts that match the user's needs, making the system serve itself in the selection process.
Solution Approach 2:
The system uses conversation history as feedback to continuously improve plug-in and prompt recommendations. By analyzing the sequence of user queries and AI responses, the system learns from the interaction pattern and provides increasingly accurate recommendations, reducing selection time while maintaining ease of use.
2Adaptability or versatility
If the number of plug-ins is increased to expand service range, then more functions are available to users, but users have difficulty knowing which plug-ins exist and selecting appropriate ones
Solution Approach 1:
The system extracts only the relevant plug-ins and prompts from the entire available set based on the current conversation context. Instead of presenting all available plug-ins to users, the AI identifies and extracts specifically those that are relevant to the user's current needs, simplifying the selection process while maintaining access to the full range of services.
Solution Approach 2:
The system applies different levels of recommendation based on local conversation context. Rather than treating all plug-ins uniformly, the AI analyzes the specific situation and provides tailored recommendations that are locally optimized for the current conversational need, making the complex set of plug-ins manageable through context-aware filtering.
3Productivity
If registered prompts and plug-ins are provided for users, then users can conveniently query required matters, but users still face difficulty in selecting appropriate registered prompts or plug-ins from the plurality available
Solution Approach 1:
The system performs preliminary analysis of the conversation context to pre-identify suitable plug-ins and prompts before the user needs to make a selection. By anticipating user needs based on conversation history and pre-filtering the available options, the system reduces the cognitive load on users and makes the selection process more efficient while maintaining ease of operation.
Data Source
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.