Robot-to-Human Transfer Model for Customer Service
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot customer service systems often fail to provide satisfactory answers to unconventional customer questions, leading to inefficient operation and lower customer satisfaction in customer service centers, as the transition from robot to human intervention is not accurately determined.
Innovation Solution
A method and apparatus that utilize a confidence score evaluation model, combining linear and deep neural network sub-models, to assess conversation and state characteristics of customers, determining the optimal time for human intervention based on machine learning algorithms and Natural Language Processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If robot customer service is used to handle all customer questions, then operation efficiency is improved and human representatives are released, but customer satisfaction deteriorates when unconventional questions cannot be answered
Solution Approach 1:
The patent introduces a transfer judgment model as an intermediary between the robot customer service and human customer service. This model evaluates conversation characteristics and robot response characteristics to determine the optimal transfer timing, ensuring that customers are transferred to human representatives only when necessary, thus maintaining high operation efficiency while preventing customer satisfaction deterioration.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring conversation characteristics and robot response characteristics during the customer service interaction. This feedback is used by the transfer judgment model to dynamically adjust transfer decisions, allowing the system to learn from past interactions and improve both efficiency and satisfaction over time.
2Reliability
If transfer to human customer service occurs frequently, then customer satisfaction is maintained, but operation efficiency deteriorates due to excessive human intervention
Solution Approach 1:
The transfer judgment model serves as an intelligent intermediary that objectively evaluates whether transfer to human service is truly necessary. By analyzing both conversation characteristics and robot response characteristics, the model prevents unnecessary transfers, ensuring that human representatives are only engaged when the robot genuinely cannot satisfy the customer, thus maintaining operation efficiency while protecting customer satisfaction.
3Productivity
If transfer timing is delayed, then robot handles more questions improving efficiency, but customer satisfaction worsens when the robot cannot answer unconventional questions
Solution Approach 1:
The transfer judgment model acts as a real-time intermediary that continuously assesses the conversation state. It analyzes conversation characteristics (such as conversation length, topic changes) and robot response characteristics (such as response confidence, multiple attempts) to identify the optimal transfer moment, ensuring timely intervention when the robot encounters unconventional questions while maximizing the use of automated service for routine matters.
Solution Approach 2:
The system performs preliminary evaluation of transfer necessity by monitoring conversation and response characteristics throughout the interaction. This allows the system to prepare for potential transfer decisions in advance, identifying patterns that indicate upcoming customer frustration or inability to resolve issues, thereby enabling proactive rather than reactive transfer decisions.
4Reliability
If transfer timing is advanced, then customer satisfaction is maintained, but operation efficiency worsens due to excessive human intervention
Solution Approach 1:
The transfer judgment model serves as an intelligent filter that objectively determines whether early transfer is truly warranted. By evaluating multiple characteristics including conversation progress and robot confidence levels, the model prevents premature transfers for routine matters while allowing early transfer when genuine issues arise, thus balancing customer satisfaction with operational efficiency.
Data Source
AI summary
Methods, systems, and devices, including computer programs encoded on computer storage media for transferring a robot customer service to a human customer service are provided. One of the methods includes: obtaining conversation characteristics from at least one round of conversations between the robot customer service and a customer; obtaining state characteristics of the customer; inputting the conversation characteristics and the state characteristics into a confidence score evaluation model to obtain a confidence score evaluation value; and when the confidence score evaluation value meets a robot-to-human intervention condition, transferring the customer to the human customer service. The confidence score evaluation model is a machine learning model, comprising a linear sub-model input with the conversation characteristics and a deep neural network sub-model input with the state characteristics.