Chatbot Prompt Engineering with Conversation Flowcharts

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current chatbots rely on predetermined scripts and conversation flowcharts, limiting flexibility and personalization, resulting in robotic and unsatisfactory customer interactions.

Innovation Solution

The method involves generating prompts for a large language model based on user input and a conversation flowchart, using a generative pre-trained transformer network to produce natural and human-like responses that adapt to individual customers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If predetermined scripts and conversation flowcharts are used for chatbots, then reliability and consistency of responses are improved, but adaptability and personalization deteriorate

Engineering Contradiction:

Improveresponse consistencyVSAvoidpersonalization

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent merges the structured reliability of conversation flowcharts with the adaptive capabilities of large language models. The system combines both approaches by using flowcharts for task routing and LLMs for natural language generation, achieving both consistency and personalization simultaneously

Inventive Principle:

Principle #5Merging (Combining)

2Ease of manufacture

If predetermined scripts are used for chatbot responses, then ease of manufacture and implementation are improved, but adaptability to individual customers deteriorates

Engineering Contradiction:

Improveimplementation simplicityVSAvoidcustomer adaptation

Core Design Contradiction:

Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The patent introduces dynamic response generation by replacing static predetermined scripts with large language models that can adapt to individual customers in real-time. The system maintains ease of implementation through automated prompt generation and flowchart-based task routing while achieving dynamic adaptability

Inventive Principle:

Principle #15Dynamics

3Device complexity

If traditional conversation engineering is used, then device complexity is reduced, but labor intensity and time consumption increase

Engineering Contradiction:

Improvesystem complexityVSAvoidresponse efficiency

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The patent implements self-service through automated prompt generation and self-play testing that enables the chatbot system to optimize its own performance without extensive manual conversation engineering. The system automatically generates prompts, performs testing, and iterates on responses, reducing both labor intensity and time consumption

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20240428013A1Enhancing chatbots through prompt-engineered large language models

Publication Date: 2024.12.26 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A method, computer program product, and computer system are provided for enhancing chatbot responses. Data corresponding to a user input to a chatbot is received. A prompt is generated for a large language model based on the received data and a conversation flowchart associated with the chatbot. The generated prompt is input to the large language model. A natural language response to the received data is generated based on an output from the large language model. The generated natural language response substantially corresponds to a response associated with the conversation flowchart.