Real-Time Article Recommendation for Customer Service Agents
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customer service agents face delays in resolving issues due to the time-consuming process of searching through a knowledge base for relevant articles during communication sessions, which increases response times and decreases customer satisfaction.
Innovation Solution
A cloud platform generates an interface that obtains utterances from the conversation between agents and users, applies relevance models to select and present relevant articles from a database, combining scores from query retrieval, machine-learned models, and behavioral signals to quickly provide relevant articles to agents during live chat sessions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If agents manually search through a knowledge base to identify relevant articles, then they can find potential answers to customer issues, but the process becomes time-consuming and increases response times
Solution Approach 1:
The system enables automatic article recommendation by combining multiple scoring mechanisms (query retrieval, machine learning models, and behavioral signals) that operate autonomously to identify and present relevant articles to agents, eliminating the need for manual searching while maintaining high accuracy in article selection
Solution Approach 2:
The patent replaces the manual mechanical searching process with an automated computational system that uses machine learning models and behavioral signal analysis to rapidly identify relevant articles, substituting human effort with an intelligent automated retrieval system that operates in real-time
2Loss of information
If the system provides comprehensive knowledge base articles to agents, then agents have more information to resolve issues, but the interface becomes complex and harder to navigate
Solution Approach 1:
The system applies different quality characteristics to different parts of the information presentation by providing comprehensive article content while using a simplified interface that highlights only the most relevant portions through targeted recommendations based on behavioral signals and machine learning scores, allowing agents to access complete information without being overwhelmed by interface complexity
3Measurement precision
If the system uses multiple relevance models and behavioral signals to select articles, then the accuracy of article recommendation improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent segments the relevance scoring process into distinct independent components (query retrieval model, machine learning relevance model, and behavioral signal analysis) that can be computed separately and then combined, allowing each component to be optimized independently while maintaining overall system accuracy and managing computational complexity through modular architecture
Data Source
AI summary
A cloud platform establishes a communication session between an agent and a user. The communication session is over an electrical medium. The cloud platform generates an interface on a client device associated with the agent. A first portion of the interface is configured to exchange messages between the agent and the user for a conversation or otherwise transcribe a conversation between the agent and the user. The cloud platform obtains, at a first time, a set of utterances from a transcript of the conversation. The cloud platform accesses a database including a plurality of articles. The cloud platform generates relevance scores between the conversation and the plurality of articles. The cloud platform then selects a subset of articles having relevance scores above a threshold value or proportion. The identified articles are presented on a second portion of the interface.