Ticket Routing via Entropy-Based Probability Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual routing of request data in computer networks is cumbersome and prone to errors, leading to increased resolution times and resource consumption.
Innovation Solution
A method for ticket routing that involves calculating probabilities based on description data and determining an entropy value to automatically route ticket data to the most likely category, with manual input requested only when the entropy value does not satisfy a threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual routing of ticket data is performed, then routing accuracy may be maintained through human judgment, but the process becomes cumbersome and time-consuming
Solution Approach 1:
The system performs preliminary classification of ticket data by calculating probabilities and entropy values before final routing decisions. This preliminary analysis enables automated routing for clear cases while preparing information for manual review when needed, reducing overall resolution time while maintaining accuracy.
Solution Approach 2:
The patent introduces an intermediary automated classification system that acts as a mediator between ticket submission and final routing. This intermediary evaluates ticket data using entropy calculations and probability assessments, handling routine cases automatically while flagging ambiguous cases for human review, thus balancing speed and accuracy.
2Productivity
If automated routing based on simple rules is implemented, then processing speed increases, but routing accuracy decreases leading to incorrect routing
Solution Approach 1:
The system changes the parameter of routing decision-making from simple rule-based classification to probability-based classification with entropy evaluation. By calculating the probability that ticket data belongs to each category and assessing entropy values, the system achieves both automated speed and improved accuracy, routing only when confidence thresholds are met.
Solution Approach 2:
The patent implements feedback mechanisms where the system evaluates the confidence level of its routing decisions through entropy calculations. When entropy values indicate low confidence or ambiguous classification, the system feeds this information back to trigger manual review, thereby maintaining high routing accuracy while preserving automated processing for clear cases.
3Reliability
If manual review of all ticket data is performed, then routing accuracy is maximized, but resource consumption and storage burden increase
Solution Approach 1:
The system applies partial manual review action by automatically processing ticket data that meets clear classification criteria while reserving manual review only for cases with high entropy values or ambiguous probabilities. This partial application of manual review maintains routing accuracy for critical cases while avoiding the resource exhaustion that would result from reviewing all tickets manually.
Solution Approach 2:
The patent enables the routing system to serve itself by implementing automated classification and confidence assessment. The system independently evaluates ticket data, calculates entropy values, and makes routing decisions for clear cases without human intervention, thereby reducing network resource consumption and storage burden while maintaining accuracy through automated quality assessment.
Data Source
AI summary
A method includes receiving, at a processor, ticket data representing a ticket. The method further includes receiving, at the processor, description data representing a description of the ticket. The method further includes calculating, based on the description data, a first probability that the ticket corresponds to a first category and a second probability that the ticket corresponds to a second category. The method further includes determining an entropy value associated with routing the ticket data. The method further includes, in response to the entropy value satisfying a threshold and the first probability exceeding the second probability, routing the ticket data to a device associated with the first category.