Automated Self-Help System Configuration via Dialogue Tree Mining
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing self-help systems for contact centers are inefficiently configured, requiring manual and time-consuming processes for customization, and fail to adapt to actual customer behavior and preferences.
Innovation Solution
A method and system that analyze prior interactions between customers and agents to generate a dialogue tree, automatically configuring an automated self-help system by recognizing speech, clustering phrases, filtering sequences, and mining a preliminary dialogue tree, allowing for user input to customize and optimize the system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual configuration processes are used for self-help systems, then customization can be achieved, but the configuration time and effort increase significantly
Solution Approach 1:
The system performs preliminary analysis of customer-agent interactions to pre-generate dialogue trees and configuration settings. By analyzing historical interaction data beforehand, the system prepares customized self-help configurations in advance, eliminating the need for time-consuming manual setup when the system is actually deployed.
Solution Approach 2:
The system automatically configures itself by analyzing its own operational data from customer interactions. Through self-service configuration, the system extracts patterns from historical data, generates optimized dialogue trees, and updates its own settings without external intervention, dramatically reducing configuration time while maintaining high adaptability.
2Adaptability or versatility
If self-help systems are customized on a per-organization basis, then relevance to customers is improved, but the complexity of system setup increases
Solution Approach 1:
The system automatically analyzes organization-specific interaction data and generates customized configurations autonomously. By using self-service configuration, the system handles the complexity of per-organization customization internally through automated pattern recognition and dialogue tree generation, presenting a simplified interface to users while maintaining high organization-specific relevance.
Solution Approach 2:
The system performs preliminary analysis of organizational interaction patterns and pre-generates customized dialogue trees before deployment. This preliminary customization process automatically adapts the system to organization-specific needs without requiring complex manual configuration, reducing setup complexity while maintaining relevance.
3Productivity
If automated configuration is implemented, then configuration speed increases, but the need for user input and control may be reduced
Solution Approach 1:
The system incorporates feedback mechanisms that allow users to review, validate, and adjust automatically generated configurations. Users can provide feedback on the generated dialogue trees and interaction patterns, and the system uses this feedback to refine its automated configuration process, maintaining both high speed and user control.
Solution Approach 2:
The configuration system is designed to be dynamic and adaptable, allowing users to intervene at various stages of the automated configuration process. Users can adjust parameters, modify generated dialogue trees, and control the level of automation, enabling the system to adapt its degree of automation based on user needs while maintaining high configuration speed.
Data Source
AI summary
A method for configuring an automated self-help system based on prior interactions between a plurality of customers and a plurality of agents of a contact center includes: recognizing, by a processor, speech in the prior interactions between customers and agents to generate recognized text, the recognized text including a plurality of phrases, the phrases being classified into a plurality of clusters; extracting, by the processor, a plurality of sequences of clusters, each of the sequences of clusters corresponding to the phrases of one of the prior interactions; filtering, by the processor, the sequences of clusters based on a criterion; mining, by the processor, a preliminary dialog tree from the sequences of clusters; invoking configuration of the automated self-help system based on the preliminary dialog tree; and outputting a dialog tree for configuring the automated self-help system.