Vector Embedding for Source Code Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for detecting defective source code commits in microservices deployments are inefficient, often requiring manual log analysis and human intervention, which can lead to misidentification of issues and delayed resolution in production environments.
Innovation Solution
A computer-implemented method using deep learning models, specifically Siamese neural networks, to analyze log differences and system metrics, generating vector embeddings that determine the likelihood of defects in source code commits, allowing for automated identification and remediation of potentially problematic commits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual log analysis and human intervention are used to detect defective source code commits, then detection accuracy can be maintained through human judgment, but detection efficiency and speed are significantly reduced
Solution Approach 1:
The patent replaces manual human analysis of logs with an automated machine learning system that processes log data, generates embeddings, and identifies defective commits automatically. This substitution eliminates the time-consuming manual review process while maintaining detection capability through algorithmic analysis of log patterns and similarities.
Solution Approach 2:
The system creates vector embeddings as digital representations (copies) of log data, allowing automated comparison and analysis of commit patterns without requiring human reviewers to manually examine raw logs. These embeddings enable efficient automated similarity searches and defect identification.
2Ease of operation
If manual log analysis is performed to identify defective commits, then detailed human insight can be applied, but the process becomes complex and requires significant human resources
Solution Approach 1:
The system performs self-service by automatically ingesting log data, generating embeddings, training models, and identifying defective commits without requiring human operators to manually analyze logs. The automated pipeline handles the entire detection process, reducing operational complexity despite the sophisticated underlying technology.
3Reliability
If traditional detection methods are used, then implementation is simpler with fewer advanced technologies, but detection accuracy and reliability are reduced
Solution Approach 1:
The patent segments the defect detection process into distinct modular components: log ingestion, embedding generation, model training, and defect identification. This segmentation allows each component to be optimized independently while working together to achieve high detection accuracy, managing complexity through modular architecture.
Solution Approach 2:
The system transforms log data into a different dimensional representation through vector embeddings, converting unstructured text logs into structured numerical vectors that can be efficiently processed by machine learning models. This dimensional transformation enables sophisticated pattern recognition and similarity analysis for improved detection accuracy.
Data Source
AI summary
In an approach, one or more computer processors create a dictionary for each source code commit in a set of historical source code commits associated with a software deployment; create a similarity model based on the created dictionary for each source code commit in the set of historical source code commits; generate a vector embedding for a source code commit pair based on a set of log differences between source code commit pairs utilizing the created similarity model; generate, responsive to a new source code commit, a new vector embedding based on a set of log differences between the new source code commit and a preceding source code commit utilizing the created similarity model; generate a defect likelihood utilizing the generated new vector embedding; determine, responsive to the generated defect likelihood exceeding a defect likelihood threshold, that the new source code commit contains defects.