Machine Learning Support Document Linking for Missing Issue Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face inefficiencies and errors in maintaining links between causing and solving support documents, leading to manual, time-consuming, and error-prone processes, which can result in unresolved customer issues and increased costs.
Innovation Solution
A machine learning-based system automatically identifies potential solving support documents by analyzing descriptive text, generating link probabilities, and producing reports for developer review to establish or confirm missing links.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual review is used to identify missing links between support documents, then developers can ensure accuracy of link identification, but the process becomes time-consuming and error-prone
Solution Approach 1:
The patent replaces the manual mechanical review process with an automated machine learning system that uses natural language processing to analyze support document text and identify missing links. The ML model automatically scores potential links, eliminating the need for time-consuming manual review while maintaining or improving identification accuracy.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between the support documents and the link identification process. This intermediary automatically analyzes document text, generates link scores, and prioritizes potential missing links, serving as a bridge that automates what previously required direct human intervention.
2Reliability
If developers manually maintain link tables between causing and solving support documents, then link accuracy can be ensured, but the workload increases significantly with hundreds of thousands of documents
Solution Approach 1:
The patent enables the link table to maintain itself automatically through the machine learning system. The ML model continuously analyzes new support documents, identifies potential missing links, and updates the link table without requiring developer intervention. This self-maintaining system scales efficiently regardless of the number of documents.
Solution Approach 2:
The patent replaces the manual maintenance mechanism with an automated ML-based system that continuously monitors support documents and updates link relationships. This substitution eliminates the repetitive manual workload while maintaining high accuracy through automated text analysis and link scoring.
3Reliability
If automated services detect missing support document implementations, then customer issues can be identified early, but false positives increase without accurate link data
Solution Approach 1:
The patent applies partial automation by using the ML model to score and prioritize potential links before automated detection services act on them. Rather than fully automated detection without verification, the system performs partial analysis to filter and rank candidates, reducing false positives while maintaining automated detection benefits.
Solution Approach 2:
The patent implements feedback loops where the ML model's link predictions are continuously refined based on actual link table data and detection service outcomes. This feedback mechanism improves the accuracy of link identification over time, reducing false positives in customer issue detection while maintaining high reliability.
Data Source
AI summary
A support document data store contains multiple support documents for a software component (including a support document identifier and descriptive text). A missing link server automatically identifies some support documents as being potential solving support documents. For each potential solving support document, a machine learning analysis of the descriptive text is performed to generate a link probability. For each document having a link probability above a threshold, a potential link message is automatically generated that includes the document identifier and an associated causing support document identifier. According to some embodiments, potential link messages are compiled into a potential link report for review by a developer to classify them as an actual link or not an actual link. A support document link data store may include indications of causing support documents with an associated links to solving support documents.