Altran unleashed a new machine-learning based tool that helps developers predict the presence of bugs in software source code. The tool, which is called Code Detect AI, works by applying machine learning (ML) to historical data in order to identify potential areas of code that are most likely to be "buggy."
The tool suggests a set of tests to diagnose and fix the potential problems, which speeds up the development cycle while minimizing the cost of fixing bugs. Altran, which is owned by consulting firm Capgemini SE, developed the new tool by working with Microsoft. The tool is available at GitHub.
Altran said that bugs were a fact of life for software development and the later they're detected in the development cycle the more the cost increases to fix them. Code Defect AI employs various ML techniques including random decision forests, support vector machines, multilayer perception (MLP) and logistic regression.
Historical data is extracted, pre-processed and labeled to train the algorithm and curate a reliable decision model. The developers are also given a confidence score that predicts whether the code is compliant or presents the risk of containing bug.
“It’s well-known that software developers are under constant pressure to release code fast without compromising on quality,” said Walid Negm, group chief innovation officer at Altran, in a statement. “The reality, however, is that the software release cycle needs more than automation of assembly and delivery activities. It needs algorithms that can help make strategic judgments ‒ especially as code gets more complex. Code Defect AI does exactly that.”
RELATED: Microsoft, AT&T create edge compute zones
Code Defect AI can be hosted on premise as well as on cloud computing platforms such as Microsoft Azure. While the solution currently supports GitHub, which is owned by Microsoft, it can be integrated with other source-code management tools as needed.
The tool is on the Microsoft AI Lab portal so that Microsoft developers can download the solution and use it internally.