Probabilistic Classification Model for Request Handler Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated assignment of electronic requests to request handlers often results in incorrect assignments, leading to wasted resources and delayed completion, as existing systems lack efficient methods to match requests with the most suitable handlers based on their experience and skill sets.
Innovation Solution
A computing device identifies previous electronic requests using structured data fields, trains a probabilistic classification model, and executes it using unstructured data fields to select a request handler with the best match in prior experience and skill set, employing probabilistic, neural network, or pre-trained classification models based on available handlers and previous requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated assignment systems are used to assign electronic requests to request handlers, then productivity is improved through automation, but assignment accuracy deteriorates leading to incorrect assignments
Solution Approach 1:
The patent introduces a classification model as an intermediary between the automated assignment system and request handlers. This model analyzes request characteristics and handler profiles to generate matching scores, serving as a mediator that improves assignment accuracy while maintaining automation efficiency. The classification model acts as a bridge that translates automated processing into accurate human-handler assignments.
Solution Approach 2:
The system implements feedback mechanisms where assignment outcomes are continuously monitored and used to retrain and improve the classification model. Correct and incorrect assignments provide feedback signals that adjust the model's parameters, enabling the system to learn from past performance and continuously improve assignment accuracy while maintaining high productivity through automation.
2Device complexity
If simple automated assignment rules are used, then device complexity is reduced, but assignment quality deteriorates resulting in wasted resources
Solution Approach 1:
The patent transforms the assignment problem from simple rule-based matching to a parameter-driven classification approach. By analyzing multiple parameters including request characteristics, handler skills, and historical performance data, the system achieves high assignment quality. The classification model dynamically adjusts parameter weights and thresholds to optimize resource allocation while maintaining reasonable system complexity.
3Reliability
If manual review of assignments is performed to ensure accuracy, then assignment quality is improved, but productivity deteriorates due to time consumption
Solution Approach 1:
The patent implements a hybrid approach where the classification model handles the majority of assignments automatically with high accuracy, eliminating the need for manual review in most cases. Manual review is reserved only for edge cases or low-confidence predictions, applying partial human intervention exactly where needed. This partial action approach maintains assignment quality while preserving productivity by avoiding unnecessary manual processing of routine assignments.
Data Source
AI summary
A computing device includes a processor and a medium storing instructions. The instructions are executable by the processor to: in response to a receipt of an electronic request comprising one or more structured data fields and one or more unstructured data fields, identify a set of previous electronic requests using the one or more structured data fields of the received electronic request; train a probabilistic classification model using at least one structured data field of the identified set of previous electronic requests; execute the trained probabilistic classification model using the one or more unstructured data fields of the received electronic request; and automatically select a request handler using an output of the executed probabilistic classification model.