Problem Statement Classifier for Human-Agent Conversation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user-agent interaction systems, particularly IVR systems, face challenges in accurately extracting user problem statements when users bypass the IVR and directly interact with human agents, leading to erroneous or irrelevant data collection.
Innovation Solution
A system and method that utilize a problem statement classifier to analyze natural language conversations between human agents and users, generating predictions on which utterances correspond to problem statements, using encoders and prediction modules to create utterance embeddings and identify relevant statements for downstream processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If IVR system prompts user to provide problem statement, then data collection is standardized and efficient, but users can bypass IVR to speak to human agents causing erroneous or irrelevant data collection
Solution Approach 1:
The patent introduces an intermediary component (problem statement classifier) that processes utterances from human-agent conversations to identify and extract problem statements. This mediator bridges the gap between unstructured human-agent interactions and the need for structured problem statement data, allowing the system to handle both IVR and direct human-agent pathways reliably.
2Measurement precision
If IVR system collects problem statements through standardized prompts, then data extraction is straightforward and reliable, but the system cannot process non-standardized human-agent conversations
Solution Approach 1:
The patent replaces the mechanical IVR prompt-based collection system with an AI-based natural language processing system. The problem statement classifier uses machine learning models to automatically identify problem statements in unstructured human-agent conversations, substituting rigid mechanical interaction with flexible intelligent processing.
Solution Approach 2:
The system changes the parameters of conversation processing from standardized prompt-response patterns to flexible natural language analysis. By adjusting the classification thresholds and model parameters, the system can adapt to different conversation styles while maintaining precision in problem statement identification.
3Quantity of substance
If the system processes all utterances in human-agent conversations, then comprehensive data is collected, but computational resources and processing time increase significantly
Solution Approach 1:
The patent extracts only the relevant problem statement utterances from the entire conversation transcript, rather than processing all utterances equally. The problem statement classifier identifies and extracts only the specific portions containing problem statements, significantly reducing the volume of data that requires detailed processing while maintaining comprehensive coverage of important information.
Data Source
AI summary
The present disclosure relates to a system, method, and computer program for extracting utterances corresponding to a user problem statement in a conversation between a human agent and a user. The system obtains a set of utterances from a natural language conversation between the human agent and the user. The system uses a problem-statement classifier to obtain machine-generated predictions as to whether each natural language utterance in the set relates to a problem statement. The system selects one or more utterances from the set as corresponding to a problem statement based on the predictions. The system provides the selected utterances to a downstream system for further processing. In certain embodiments, the problem statement classifier includes an encoder that creates an utterance embedding for each utterance and a prediction module that uses the utterance embeddings to predict whether each utterance corresponds to a user problem statement.