Predictive Contact Center Routing Using Speech and Sentiment Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional skill-based routing in contact centers is static and requires manual effort, failing to leverage new data types and adapt dynamically to real-time changes, leading to suboptimal customer-agent matching.
Innovation Solution
A method for generating predictive routing models using speech and sentiment analytics, involving data processing, feature engineering, and machine learning to correlate agent-customer pairings based on interaction outcomes, customer profiles, and agent attributes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional skill-based routing with explicit skill models is used, then routing decisions can be made based on agent skills, but the models are static and do not dynamically adapt to real-time changes
Solution Approach 1:
The patent transforms static skill-based routing models into dynamic predictive routing models that continuously adapt to real-time changes. The system uses historical interaction data, speech analytics, and sentiment analysis to create models that evolve over time, capturing changing customer preferences, agent performance patterns, and interaction outcomes without requiring manual model reconstruction.
Solution Approach 2:
The predictive routing model automatically learns and updates from historical data without requiring manual intervention for model construction or maintenance. The system self-trains on past interactions, automatically identifying patterns and relationships between agent-customer pairings and outcomes, thereby eliminating the manual effort traditionally required to refine skill models.
2Measurement precision
If refined skill models are constructed manually, then routing accuracy improves, but the cost and manual effort increase significantly
Solution Approach 1:
The system automatically trains predictive routing models using historical interaction data, eliminating the need for manual model construction. The machine learning algorithms self-learning patterns from past data, automatically refining routing accuracy without requiring manual intervention, thereby achieving high precision with minimal human effort.
Solution Approach 2:
The patent replaces manual model construction and maintenance activities with automated machine learning systems. Instead of manually refining skill models, the system uses computational algorithms to automatically train predictive models on historical data, substituting mechanical manual processes with automated intelligent systems that achieve superior accuracy.
3Adaptability or versatility
If traditional skill-based routing is used, then implementation is straightforward, but new data types such as speech and sentiment analytics are not leveraged
Solution Approach 1:
The predictive routing model serves multiple functions by integrating diverse data types including historical interaction data, speech analytics, and sentiment analysis. The system processes and synthesizes these different data sources to generate comprehensive routing decisions, making the system versatile and adaptable to various data formats while managing complexity through unified model architecture.
4Productivity
If more data processing and analysis are performed, then routing optimization improves, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary processing of historical interaction data, speech analytics, and sentiment analysis in advance to train predictive routing models. By pre-processing and pre-training on historical data, the system prepares optimized models that can make rapid routing decisions in real-time, reducing the processing time required during actual customer interactions while maintaining high optimization quality.
Data Source
AI summary
Systems and methods for generating a routing recommendation for an incoming interaction in a contact center are described. For example, the method includes receiving interaction data associated with the incoming interaction, determining candidate agents based on one or more constraints, obtaining agent data related to the candidate agents, determining, for each of the candidate agents, an expected performance score using a predictive routing model, the expected performance score indicating a predicted performance of the corresponding candidate agent relative to a predetermined performance metric for handling the incoming interaction, and generating a routing recommendation for the incoming interaction based on the expected performance scores of the candidate agents, the routing recommendation identifying one or more agents from the candidate agents predicted to achieve a predefined level of the predetermined performance metric.