Graphical Chatbot Interface with Predicted Path Hierarchy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current chatbot systems lack the ability to efficiently guide users through probable questions and responses to quickly arrive at a final response, as they do not utilize their knowledge base to directly accelerate the user's interaction.
Innovation Solution
A graphical chatbot interface is provided that displays predicted chatbot paths, allowing users to select a specific path, thereby streamlining their interaction and enabling faster access to desired responses by utilizing historical interactions and machine learning to predict probable chatbot content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional chatbot systems are used without predicted paths, then the system structure remains simple, but the interaction efficiency and speed to arrive at final responses deteriorate
Solution Approach 1:
The system performs preliminary actions by generating predicted chatbot paths in advance based on historical interactions and machine learning. These paths are prepared beforehand and presented to users, allowing them to navigate directly to likely outcomes rather than proceeding through sequential traditional chatbot turns, thereby improving interaction efficiency without requiring fundamental changes to the underlying chatbot system architecture
Solution Approach 2:
The chatbot interaction is segmented into multiple predicted paths that branch from the main conversation flow. Each path represents a probable sequence of questions and responses based on historical data. This segmentation allows users to see and select from multiple potential conversation trajectories simultaneously, improving productivity by enabling direct navigation to desired outcomes while maintaining the original system structure
2Loss of time
If predicted chatbot paths are displayed to guide users, then the speed to arrive at final responses improves, but the complexity of the interface and processing requirements worsens
Solution Approach 1:
The system creates simplified copies or representations of the chatbot's knowledge base in the form of visual predicted paths. Instead of presenting the full complexity of the underlying machine learning models and historical interaction data, the system generates accessible visual copies showing probable conversation flows. This allows users to quickly navigate to responses without exposing interface complexity, reducing time loss while keeping the interface manageable through familiar visual metaphors
Solution Approach 2:
The system adds a visual dimension to the traditional text-based chatbot interface by displaying predicted paths as graphical elements. This dimensional addition allows users to perceive multiple conversation outcomes simultaneously in space rather than sequentially in time, dramatically reducing the time to arrive at responses. The complexity is managed by rendering these paths as simple visual guides rather than full interactive interfaces
3Measurement precision
If historical interactions and machine learning are used to predict paths, then the accuracy of predicted content improves, but the computational resources and processing time worsen
Solution Approach 1:
The system performs computationally intensive machine learning predictions in advance, generating predicted chatbot paths based on historical interactions before users arrive. These predictions are pre-computed and stored as navigable paths. When users interact with the system, they receive pre-prepared accurate predictions without requiring real-time computation, thereby achieving high prediction accuracy while minimizing real-time computational resource consumption and energy usage
Solution Approach 2:
The system extracts only the essential predicted paths from the full machine learning output and presents these to users. Rather than processing and displaying all possible conversation outcomes generated by the machine learning model, the system extracts and prioritizes the most relevant predicted paths based on probability and user context. This extraction approach maintains high prediction accuracy for the displayed paths while significantly reducing computational resource requirements compared to generating and processing the complete prediction space
Data Source
AI summary
Interaction with a chatbot of a computer system is facilitated by receiving, by the chatbot, a request from a user of the chatbot, where the request establishes a user-chatbot interaction. Based at least in part on the request, a plurality of predicted chatbot paths are generated, forming a hierarchy of predicted chatbot content for the user-chatbot interaction. A graphical chatbot interface including, at least in part, the hierarchy of predicted chatbot content for the user-chatbot interaction is provided for display on an electronic device, and a selection by the user of one predicted chatbot path of the graphical chatbot interface is received, via the electronic device, as part of the user-chatbot interaction, where the providing of the graphical chatbot interface facilitates user interaction with the chatbot along the selected predicted chatbot path.