Agent-Assist Feedback Loop for Subdocument Answer Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contact-center agents face challenges in finding relevant information to address user inquiries due to turnover and the inefficiency of existing agent-assist systems, which often provide non-pertinent answers, increasing call handling time and reducing user satisfaction.
Innovation Solution
An agent-assist system that provides context-aware, subdocument-granularity recommendations by analyzing large collections of documents to identify semantically related subdocuments and collects feedback to improve the relevance of answers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If agents look up answers in documents and knowledge bases, then they can find relevant information, but call handling time increases and productivity decreases
Solution Approach 1:
The system enables self-service by automatically providing relevant information to agents without requiring manual search. The agent-assist system autonomously queries knowledge bases, retrieves relevant documents and subdocuments, and presents them to agents, eliminating the need for agents to manually look up information and thereby reducing call handling time while maintaining information access
Solution Approach 2:
The agent-assist system acts as an intermediary between agents and knowledge bases. It receives user inputs, automatically searches relevant knowledge bases, retrieves pertinent subdocuments, and delivers filtered information to agents. This intermediary function streamlines the information retrieval process, reducing time loss while ensuring agents receive accurate, relevant information
2Productivity
If agent-assist systems provide recommended answers, then information access is improved, but relevance accuracy decreases leading to non-pertinent recommendations
Solution Approach 1:
The system segments documents into smaller subdocuments at the paragraph or sentence level, allowing for more precise matching with user inputs. Instead of returning entire documents, the agent-assist system identifies and retrieves only the specific subdocuments containing relevant information, improving both retrieval efficiency and recommendation accuracy by presenting targeted content rather than generic document-level results
Solution Approach 2:
The system changes the granularity parameter of information retrieval from document-level to subdocument-level. By adjusting this parameter, the system can retrieve more precise, relevant information segments that closely match user queries, thereby improving recommendation accuracy while maintaining efficient information access
3Reliability
If extensive training is provided to agents, then domain expertise improves, but training time and costs increase
Solution Approach 1:
The system provides self-service knowledge delivery by automatically presenting relevant subdocuments and information to agents during interactions. This on-demand knowledge provision allows agents to access expert-level information without requiring extensive formal training, effectively transferring domain expertise through the system rather than through lengthy training programs
Solution Approach 2:
The system performs preliminary action by pre-processing and organizing knowledge base content into searchable subdocuments before agents need them. Relevant information is prepared and structured in advance, allowing agents to quickly access domain expertise during interactions without requiring extensive prior training to navigate or interpret complex information
Data Source
AI summary
Techniques for agent-assist systems to provide context-aware, subdocument-granularity recommended answers to agents that are attempting to answer queries of users. The agent-assist system may obtain collections of documents that include information for responding to queries, and analyze those documents to identify subdocuments that are associated with different semantics or meanings. Subsequently, any queries received can be analyzed to identify their semantics, and relevant subdocuments can be identified as having similar semantics. When the agent-assist system presents the agent with the relevant documents, it may highlight or otherwise indicate the relevant subdocument within the document for quick identification by the agent. Further, the agent-assist system may collect feedback from the agent and/or user to determine a relevancy of the recommended answers. The agent-assist system can use the feedback to improve the quality of the recommended answers provided to the agents.