Automated Abnormal Event Diagnosis Using Decision Tree Weights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users often face difficulties in describing technical issues to customer service representatives, leading to increased communication time and labor costs due to misunderstandings and unclear problem descriptions.
Innovation Solution
A data processing method and apparatus that automatically detect abnormal events through interaction data, obtain user attribute information, and utilize a decision tree model with weighted correspondences to determine the cause of issues without user-customer service representative interaction, allowing for automated problem-solving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If customer service representatives manually determine the cause of abnormal events through communication with users, then they can understand the problem, but communication time increases and labor costs increase
Solution Approach 1:
The system enables automated self-diagnosis by having the terminal device automatically collect interaction data, user attribute information, and abnormal event information, then send this data to the server for automatic cause determination using a decision tree model, eliminating the need for customer service representatives to manually communicate with users
Solution Approach 2:
The manual communication and diagnostic process between customer service representatives and users is replaced with an automated electronic system that collects data, processes it through a decision tree model on the server, and automatically determines the cause of abnormal events
2Reliability
If customer service representatives manually determine the cause of abnormal events, then they can provide help, but labor costs increase
Solution Approach 1:
The system performs automated cause determination and generates processing suggestions without requiring customer service representative intervention, allowing the system to serve itself in diagnosing and resolving abnormal events
Solution Approach 2:
A decision tree model serves as an intermediary between the collected data and the cause determination, automatically processing interaction data, user attribute information, and abnormal event information to identify causes and generate processing suggestions
3Loss of information
If users describe problems to customer service representatives, then the problem can be communicated, but misunderstandings occur due to different perspectives
Solution Approach 1:
The verbal communication process between users and customer service representatives is replaced with automated data collection that directly extracts interaction data, user attribute information, and abnormal event information from the terminal device and sends it to the server for processing
Solution Approach 2:
The decision tree model acts as an intermediary that processes the collected data objectively without the perspective differences that occur in human communication, automatically determining the cause based on the structured data
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Implementations of the present invention provide a data processing method and apparatus. The method includes: obtaining interaction data between a user and a server; detecting, based on the interaction data, whether an abnormal event occurred; when an abnormal event occurred, obtaining attribute information of the user; searching a locally stored first correspondence between attribute information corresponding to the abnormal event, a cause, and a weight for a record that includes the attribute information and a largest weight; and determining a cause in the record as a cause of the abnormal event. In this whole process, the user does not need to communicate with a customer service representative, the customer service representative does not need to determine the cause of the abnormal event by communicating with the user, and the cause of the abnormal event can be determined by using the attribute information of the user and the first correspondence between the attribute information corresponding to the abnormal event, the cause, and the weight. Therefore, labor costs are saved.