Robot-to-Human Chat Switching Through Predictive Sentiment Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customer dissatisfaction arises when customer service systems rely solely on robot chat due to understanding limitations, while human service is costly, necessitating accurate timing for switching between robot and human chat to optimize efficiency and cost.
Innovation Solution
A method and system using AI/ML-based sentiment detection to predict user sentiment from current, historical, and predicted discourse, switching from robot to human chat when negative sentiment is detected, employing LSTM neural networks and BERT for discourse analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If robot chat is always used to serve customers, then labor cost is reduced, but user satisfaction deteriorates due to robot's understanding limitations
Solution Approach 1:
The system dynamically switches between robot chat and human chat based on real-time sentiment analysis. When user sentiment becomes negative, the system transitions from robot to human chat, and vice versa. This dynamic adaptation resolves the contradiction by allowing the system to leverage robot efficiency for positive sentiment interactions while ensuring user satisfaction through human intervention when needed.
Solution Approach 2:
The system monitors sentiment scores as a key parameter and uses threshold-based decision making to determine chat object switching. By changing the operational parameter (chat object type) based on sentiment parameter changes, the system optimizes both labor cost and user satisfaction.
2Reliability
If human service is always used as the virtual chat object, then user satisfaction is improved, but labor cost increases
Solution Approach 1:
Instead of always using human service, the system applies human service partially - only when sentiment analysis indicates negative user sentiment. For positive or neutral sentiment cases, robot chat suffices. This partial application of human service maintains user satisfaction while significantly reducing labor costs compared to always using human agents.
Solution Approach 2:
The robot chat system serves itself by handling routine customer service tasks autonomously. Through sentiment analysis, the system self-regulates when to escalate to human chat, reducing the need for continuous human monitoring and intervention, thereby lowering labor costs while maintaining service quality.
3Device complexity
If robot chat is used without sentiment detection, then system complexity is reduced, but communication efficiency deteriorates
Solution Approach 1:
The system implements feedback through sentiment analysis, where user responses are continuously analyzed and fed back to determine subsequent chat object selection. This feedback mechanism enables the system to adapt to user needs in real-time, significantly improving communication efficiency despite the added complexity of sentiment detection algorithms.
4Measurement precision
If sentiment analysis is implemented to predict switching timing, then switching accuracy is improved, but computational resources increase
Solution Approach 1:
The system performs preliminary sentiment analysis on user inputs before determining chat object switching. By analyzing sentiment in advance and predicting future user needs, the system can proactively switch chat objects before user dissatisfaction escalates, improving switching accuracy while allowing for optimized resource allocation through predictive rather than reactive processing.
Data Source
AI summary
Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for switching a chat object. The method includes: using a robot as a chat object to converse with a user; outputting a target sentiment score based on current discourse, historical discourse, and predicted discourse associated with the user; and switching the chat object conversing with the user from the robot to human in response to the target sentiment score indicating that the user's sentiment is a negative sentiment. Implementation of this method can improve human-machine interactivity and communication efficiency, and ensure the sustainability of communication between a chat system and a user, thus improving the user's satisfaction with a service system.