AI Support Assistant Using Messaging Logs for Faster Issue Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Support agents face challenges in quickly locating accurate information to resolve user issues due to inefficient search methods in existing systems, leading to resource wastage and prolonged issue resolution times, especially in complex service environments.
Innovation Solution
A machine learning model trained on historical messaging platform communications provides real-time answers and generates practice queries to assist support agents, reducing manual searches and enhancing training efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If support agents manually search policy databases and chat logs to locate relevant information, then they can find answers to user issues, but the process is slow and consumes excessive computing resources and time
Solution Approach 1:
The system pre-processes and indexes historical chat logs and policy documents before they are needed for support operations. By performing indexing and structuring of data in advance, the system enables rapid retrieval during actual support interactions, eliminating the need for agents to manually search through unorganized data.
Solution Approach 2:
The patent introduces an intermediary system (chatbot or virtual assistant) that acts as a mediator between support agents and the knowledge base. This intermediary automatically queries, retrieves, and presents relevant information from policy databases and chat logs, freeing agents from manual searching while maintaining high accuracy in information delivery.
2Reliability
If support agents manually search through chat logs and policy databases, then they can locate relevant information, but computing resource consumption and latency increase significantly
Solution Approach 1:
The system performs preliminary indexing and structuring of chat logs and policy documents before they are needed for support operations. By pre-processing data into searchable formats with metadata tags and structured schemas, the system enables efficient queries that consume minimal computing resources during actual support interactions while maintaining high retrieval accuracy.
Solution Approach 2:
The patent creates simplified copies or representations of the knowledge base (such as indexed versions, summarized extracts, or pre-processed data structures) that can be queried efficiently. These copies maintain the essential information needed for accurate retrieval while requiring significantly fewer computing resources to search compared to the original unprocessed data sets.
3Adaptability or versatility
If complex services are supported with large numbers of support agents, then comprehensive client support is provided, but the complexity of managing information access and training increases
Solution Approach 1:
The patent implements a universal information access system that serves multiple functions: it acts as a knowledge base for support agents, a training resource for new agents, and a self-service tool for customers. This multi-functional system handles diverse service types and query complexities through a single unified interface, reducing the need for separate systems for each function and simplifying overall system management.
Solution Approach 2:
The intermediary system (chatbot/virtual assistant) simplifies complex information access by providing a unified, user-friendly interface that handles diverse queries across multiple service types. This mediator translates various user needs into appropriate database queries and presents information in consistent formats, reducing the complexity that support agents would otherwise face when accessing information for different complex services.
Data Source
AI summary
Artificial intelligence trained with messaging communications can be used to provide support for software services. In an example, a computing system can receive, via a graphical user interface (GUI), a first query associated with a first request from a client with respect to an issue with a service. The first query can be provided in a natural language format to a machine learning model that is trained on historical communication logs from a messaging platform. The machine learning model can generate a first output indicating a first answer to the first query. The machine learning model can also generate a second output indicating a second query associated with a second answer provided as a second input. The first answer and the second query can be presented in the natural language format on the user device via the GUI for use in resolving the issue with the service.