Hybrid IVR System with Human Agent Feedback Loop

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current Wizard of Oz systems for creating interactive voice response (IVR) systems are inefficient as they require building a human-operated system from scratch, deploy human and automated systems sequentially, and lack direct comparison of human and automated decision-making, making it difficult to identify problematic interactions and improve automated systems effectively.

Innovation Solution

A computer-implemented method that involves receiving statements at an automated response system, determining and implementing responses, storing data, and selectively identifying it based on agent input, allowing for the integration of human agent interpretations to update and improve automated responses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a human operated system is built from the ground up to begin the Wizard of Oz process, then the system can handle caller interactions, but the development time and complexity increase significantly

Engineering Contradiction:
Improvesystem functionalityVSAvoiddevelopment time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by deploying the automated system skeleton and infrastructure before human agents are fully integrated. The system framework, including call routing, speech recognition, and data collection mechanisms, is prepared in advance, allowing human agents to be added incrementally without requiring complete system reconstruction.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system is segmented into modular components: automated response modules, human agent interfaces, data collection layers, and analysis systems. This segmentation allows parallel development of different modules and enables the system to function partially with automated components while human agents are being integrated, reducing overall development time.

Inventive Principle:
Principle #1Segmentation

2Ease of manufacture

If human and automated systems are deployed in sequence, then implementation is simpler, but direct comparison of decision-making between human and automated systems is not possible

Engineering Contradiction:
Improveimplementation simplicityVSAvoiddecision-making comparison data
Core Design Contradiction:
Ease of manufactureVSLoss of information

Solution Approach 1:

The patent merges human and automated systems into a single hybrid operational framework where both systems process identical caller interactions simultaneously. The human agent interface and automated response system share the same call data, enabling direct comparison of their decision-making processes while maintaining implementation simplicity through unified data collection.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system implements feedback mechanisms that capture and store decision-making data from both human agents and automated systems in real-time. This feedback loop allows continuous comparison and analysis of decision patterns, providing valuable information for improving automated system performance while maintaining simple parallel deployment.

Inventive Principle:
Principle #23Feedback

3Quantity of substance

If data from caller interactions is collected during human system operation, then training data is available, but sorting through the data to identify problematic interactions becomes difficult

Engineering Contradiction:
Improvetraining data volumeVSAvoiddata processing complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent replaces manual data sorting and analysis with automated computational systems. Machine learning algorithms and data mining tools automatically process collected interaction data, identifying problematic patterns and training opportunities without requiring manual review of each data point, thus reducing data processing complexity while maintaining large data volumes.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system introduces intermediary data processing layers that automatically filter, categorize, and prioritize collected interaction data. These intermediary systems use automated rules and algorithms to identify problematic interactions and prepare training datasets, reducing the complexity of data sorting while preserving the quantity of available training data.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS8027457B1Process for automated deployment of natural language
Publication Date: 2011.09.27 CMG UTAH INC
  • US8027457B1 patent drawing
  • US8027457B1 patent drawing
  • US8027457B1 patent drawing

AI summary

There is disclosed a system and method for monitoring and updating an interactive voice response system through the use of a human agent who reviews the responses provided by a computer system and makes recommendations which are then used as data for updating the interactive voice response system. There is further disclosed a user interface for allowing the human agent to review the responses provided by the computer system, and to provide alternate responses which could be implemented by the computer system.