Intent Classifier for Customer Support Automation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing customer support systems are inefficient and costly, as users often face time-consuming navigation through websites, apps, or voice menus, and companies incur significant expenses hiring customer service representatives to manually handle requests.

Innovation Solution

Implementing an intent classifier to determine when automated processing is available, allowing seamless transitions between human customer service representatives and automated systems to efficiently resolve customer issues.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated processing is implemented for customer support, then operational costs are reduced and efficiency is improved, but the system complexity increases and requires intent classification capabilities

Engineering Contradiction:
Improvecustomer support efficiencyVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

An intent classifier is introduced as an intermediary component between the customer support interface and the automated processing system. The intent classifier analyzes customer communications to determine whether they represent support requests that can be handled automatically, thereby mediating between human users and the automated system while managing complexity through specialized classification logic

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system enables self-service automated processing for customer support tasks by implementing intent recognition capabilities that allow the system to autonomously identify and handle support requests without human intervention, reducing operational costs while maintaining service quality through intelligent classification

Inventive Principle:
Principle #25Self-service

2Reliability

If human customer service representatives are used, then complex issues can be resolved with high accuracy, but operational costs increase significantly

Engineering Contradiction:
Improveissue resolution accuracyVSAvoidoperational cost
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The customer support system is segmented into different handling paths based on intent classification. Routine and straightforward support requests are routed to automated processing to reduce costs, while complex or ambiguous requests are directed to human representatives, thereby optimizing the balance between accuracy and operational cost through selective task allocation

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes the operational parameter from purely human-based support to a hybrid model by introducing intent classification as a filtering mechanism. This parameter change enables the system to dynamically adjust between automated and human handling based on the characteristics of each support request, reducing overall operational costs while maintaining high accuracy for complex issues

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If users navigate through websites or voice menus to reach support, then comprehensive service information is accessible, but user time is consumed

Engineering Contradiction:
Improveservice information accessibilityVSAvoiduser navigation time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The intent classifier performs preliminary analysis of customer communications to identify support requests before users need to navigate through multiple pages or voice menus. By pre-processing and classifying the intent of user messages, the system can directly route users to appropriate automated assistance or representative connections, eliminating time-consuming navigation while maintaining comprehensive service accessibility

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3673421B1Automating communications using an intent classifier
Publication Date: 2025.05.07 ASAPP INC
  • EP3673421B1 patent drawingFigure 1A
  • EP3673421B1 patent drawingFigure 1B
  • EP3673421B1 patent drawingFigure 1C

AI summary

An intent classifier may be used to increase the efficiency of a communications system. A company may provide assistance to a first user using automated processing or by a second user manually responding to the first user. To reduce costs, the company may prefer to use automated processing for assistance where it is available. While a second user is assisting a user, a message from the first user may be processed with an intent classifier to determine that automated support is available to assist the first user, and a suggestion may be presented to the second user to transfer the first user to the automated processing, such as by presenting a button to the second user to transfer the first user to the automated processing. The second user may then transfer the first user to the automated processing and assist other users.