Predictive Analysis System for Customer Service Task Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customer service systems face challenges in providing timely and efficient support across various communication channels, especially when users contact on behalf of others, requiring systems to accurately identify tasks and criticality levels in real-time.
Innovation Solution
A predictive analysis system that uses neural networks to analyze communication data, identifying matched tasks and criticality levels by processing caller utterances, relationship data, and historical information, and generating predictive analysis results to assist agents in prioritizing user requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If customer service is provided via multiple communication channels (audio call, electronic messaging) to increase user accessibility, then user access to agents is improved, but system complexity increases
Solution Approach 1:
The system implements a universal customer service platform that handles multiple communication channels (audio calls, electronic messaging, chat) through a single integrated architecture. The server system can route different types of communications through unified processing logic, allowing agents to serve customers across various channels without requiring separate systems for each channel.
2Loss of time
If the system processes communication data in real-time to identify tasks and criticality levels, then service timeliness is improved, but computational resources increase
Solution Approach 1:
The system performs preliminary processing of communication data by extracting key features and identifying potential tasks before full analysis. The server system pre-processes incoming communications to identify urgent patterns or task indicators, allowing for faster response times without requiring full computational resources to be allocated continuously to every communication.
3Measurement precision
If the system accurately identifies tasks and criticality levels by analyzing communication data, then service precision is improved, but analysis complexity increases
Solution Approach 1:
The system segments the communication data analysis into distinct processing stages: initial data extraction, task identification, criticality assessment, and priority assignment. Each stage handles a specific aspect of the analysis independently, allowing the system to achieve high accuracy in task identification without requiring all analysis components to operate simultaneously at full complexity.
Data Source
AI summary
Method starts with processing, by a processor, audio signal to generate audio caller utterance and transcribed caller utterance. Processor generates identified tasks based on transcribed caller utterances. Processor obtains member context associated with member identification and obtains available tasks associated with member identification. Processor determines lengths of time associated with available tasks, respectively. Lengths of time associated with available tasks are lengths of time that the available tasks have been available. Processor determines messaging data associated with available tasks, respectively, that is based on dates of receipt of messages pertaining to available tasks. Processor, using a predictive analysis neural network, generates a predictive analysis result that is based on the available tasks, the lengths of time associated with the available tasks, or messaging data associated with the available tasks. Predictive analysis result includes at least one matched task that is relevant to caller. Other embodiments are disclosed herein.