Adaptive Issue Type Identification Through Capability-Based Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional issue-tracking tools require users to manually select an issue type at creation, which can lead to operational capability restrictions and inefficient workflows due to incorrect type identification, limiting access to desired capabilities.
Innovation Solution
An adaptive issue type identification system that uses a predictive machine learning model to classify issue types based on selected operational capabilities, dynamically ranking and filtering capabilities based on historical usage, and associating the appropriate issue type identifier with the issue data object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If users manually select issue type at creation, then issue type identification can be performed, but operational capability restrictions occur and workflow efficiency decreases due to incorrect type identification
Solution Approach 1:
The system enables self-service by automatically determining issue type through machine learning classification based on issue description and selected operational capabilities, eliminating the need for manual user selection and reducing errors from incorrect type identification
Solution Approach 2:
The manual mechanical process of user selection is replaced with an automated machine learning system that classifies issue types based on analyzed features, improving both accuracy and workflow efficiency
2Adaptability or versatility
If all operational capabilities are displayed to users, then complete capability selection is possible, but computational load and interface complexity increase
Solution Approach 1:
The system performs preliminary actions by pre-ranking and filtering operational capabilities based on historical usage data and relevance scoring before presentation to users, allowing complete capability access while reducing interface complexity
Solution Approach 2:
Different operational capabilities are presented with different levels of prominence based on their relevance scores and historical usage, with highly relevant capabilities displayed more prominently while less relevant ones remain accessible but less prominent
3Measurement precision
If predictive machine learning model is used for issue type classification, then classification accuracy improves, but computational resources required increase
Solution Approach 1:
The system applies partial action by using a two-stage approach: first filtering capabilities based on historical usage, then applying machine learning only to rank and select from the filtered subset, reducing overall computational load while maintaining accuracy
Solution Approach 2:
Historical usage data is analyzed in advance to pre-filter and pre-rank operational capabilities, reducing the dataset that requires computationally intensive machine learning processing
Data Source
AI summary
Embodiments provide an adaptive issue type identification platform for automatically determining an issue type in an enterprise-level software development issue-tracking application. Embodiments include receiving a request to generate an issue data object, causing initializing of the issue data object in a data store, causing display of an issue management capability selection interface, and receiving issue capability selection input in response to user engagement with the issue management capability selection interface. In response to receiving the issue capability selection input, the adaptive issue type identification platform fetches an operational capability set from a capability registry based on the issue capability selection input, causes the association of the operational capability set to the issue data object at the data store, determines an issue type identifier for the issue data object based on the operational capability set, and updates metadata associated with the issue data object to include the issue type identifier.