Bug prediction based on your code history
summary
The talk presents building a non-static (dynamic) code analyzer that uses machine learning to predict unobvious bugs from a project’s code history. It covers using Weka for exploratory ML development and explains classification algorithms—Decision Trees, SVM, and Naive Bayes—and shows real-life examples of improving the development lifecycle with these AI techniques.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 2:23Automation practices
- 7:47Hidden bug detection
- 10:37Prediction hypothesis
- 13:30Tool integration
- 20:53Data prep and ML
- 25:13ML for commit analysis
- 37:20Results and conclusions
- 40:13Q&A
related talks
38mApplying machine learning to software development to reduce bugs
47mBug prediction based on your code history
18mPredictive Testing in JavaScript with Machine Learning
35mBetter C++ using Machine Learning on Large Projects
50mPractical Applications of ML in Software Testing
37mForecasting using machine learning methods
37mMachine Learning on Source Code
45mApplying Machine Learning Methods to Static Analysis of Executables