Intent Association for Help Desk Response Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for responding to help desk messages are time-consuming and costly due to the need for manual processing and lack of efficient methods to associate user intents with agent responses, leading to inefficiencies in message handling.
Innovation Solution
A computer-implemented method and system that implicitly associates user intents with agent-selected responses by receiving and processing input text messages, computing intents, prompting agents for association, and recording responses, allowing for automated and pre-written response suggestions based on intent recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual processing of help desk messages is used, then quality of interactions is maintained, but time consumption and costs increase
Solution Approach 1:
The system performs preliminary action by pre-computing intents for incoming messages and preparing suggested responses before the agent needs to respond. The intent computation and response suggestion generation occur in advance, allowing the agent to quickly select and customize responses without manual processing delays.
Solution Approach 2:
The system enables self-service by automatically generating responses based on computed intents without requiring manual intervention for each response. The AI system serves itself by autonomously creating response suggestions that agents can then customize or send directly, reducing the manual processing burden.
2Productivity
If automated response generation is implemented, then response time decreases, but accuracy of intent classification may worsen
Solution Approach 1:
The system implements feedback by allowing agents to review, correct, and validate the computed intents and suggested responses before finalizing the response. This feedback loop ensures that intent classification accuracy is maintained while still benefiting from automated response generation, as agents can correct any misclassifications.
Solution Approach 2:
The system applies partial action by providing agents with suggested responses rather than fully automated responses, allowing agents to review and customize as needed. This partial automation approach maintains high accuracy by leveraging agent expertise while still reducing time consumption through pre-generated suggestions.
3Productivity
If pre-written responses are used for common queries, then response time decreases, but adaptability to unique situations may worsen
Solution Approach 1:
The system introduces dynamics by allowing agents to customize and modify pre-written responses based on the specific situation and user context. Rather than using static pre-written responses, the system enables dynamic adjustment of responses while maintaining the efficiency benefits of having pre-prepared response templates for common queries.
Data Source
AI summary
Apparatuses, methods, and systems for implicitly associating user intents to agent selected responses. One method includes receiving, by a server, a first input text message from a user, displaying, by the server, the first input text message to an agent, receiving, by the server, a first response to the first input text message from the agent, computing a first intent with the first input text message, prompting, by the server, the agent whether to associate the first response with the computed first intent of the first input text message, and recording the first response with the computed first intent of the first input message when the agent indicates an association between the first response with the computed first intent of the first input message.