Conversational Experience Previewing with Probabilistic Flow Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current conversational experience previewing tools require manual user interaction to generate inputs, are inefficient for large conversational journeys, and lack integration with external data to ensure comprehensive path exploration, especially for specific journey portions or most-used branches.
Innovation Solution
A method that extracts constraints and flow data from a conversational model to construct a directed graph, populates edges with probabilities, and generates a self-contained experience preview by traversing the graph, synthesizing user inputs to demonstrate user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual user interaction is used to generate inputs for conversational experience previewing, then the preview can be generated, but the process is inefficient and time-consuming especially for large conversational journeys
Solution Approach 1:
The system performs preliminary actions by automatically generating synthetic user inputs and traversing the conversational journey before manual intervention is needed. The graph construction and flow data extraction are prepared in advance, enabling automated preview generation without waiting for manual input creation.
Solution Approach 2:
The system serves itself by automatically generating the inputs needed for previewing. The synthetic data generation component creates realistic user inputs without human intervention, and the graph traversal automatically executes the conversational flow, making the system self-sufficient for preview generation.
2Reliability
If comprehensive path exploration is performed in large conversational journeys, then all possible user experiences are covered, but the complexity and resource requirements increase significantly
Solution Approach 1:
The conversational journey is segmented into discrete conversational steps that form nodes in a directed graph. Each step is independently analyzed and connected through edges representing possible transitions. This segmentation allows comprehensive exploration to be broken down into manageable units that can be processed systematically.
Solution Approach 2:
A directed graph serves as an intermediary structure between the conversational model and the preview generation process. The graph abstracts the complex conversational flow into a visualizable structure with nodes and edges, making it easier to traverse and analyze while maintaining the complete path information.
3Productivity
If flow data with probabilities is extracted and used in graph traversal, then the most-used branches can be identified efficiently, but additional data processing is required
Solution Approach 1:
Flow data extraction and probability calculation are performed as preliminary actions before the graph traversal begins. By pre-computing the probabilities and populating the graph edges with this data, the system avoids the need for complex real-time calculations during traversal, improving overall efficiency.
Data Source
AI summary
A plurality of constraints associated with conversational steps implemented by a conversation model is extracted from the conversation model. Using the conversational steps and the constraints, a directed graph is constructed, each node in the directed graph representing a conversational step, each directed edge in the directed graph representing a possible execution path from a first conversational step to a second conversational step. An edge in the graph is populated with flow data denoting a probability associated with the edge. By traversing a portion of the graph, an experience preview is generated, the experience preview demonstrating a user experience of a portion of the conversation model.