Veracode, Inc.

Asankhaya Sharma

Yaqin Zhou

Darius Tsien Wei Foo

Jie Shun Yeo

Ming Yi Ang

Hao Xiao

Hendy Heng Lee Chua

With invocations of a software development pipeline, organization specific remediations/fixes for a software project can be learned from scanning results of code submissions (e.g., commits or merges) across an organization for a software project(s). Fixes of detected program code flaws can be detected and/or specified across scans and associated with flaw identifiers and used for training machine learning models to identify candidate fixes for detected flaws. This ongoing learning during development propagates fixes created or chosen by experts (e.g., software engineers working on the software project) relevant to the software project. The experts can choose from suggestions mined from the learned fixes of the organization and suggestions generated from a pipeline created with the trained machine learning models. The selections are then used for further training of the machine learning models that form the pipeline.

Development pipeline integrated ongoing learning for assisted code remediation

A system to create a stacked classifier model combination or classifier ensemble has been designed for identification of undisclosed flaws in software components on a large-scale. This classifier ensemble is capable of at least a 54.55% improvement in precision. The system uses a K-folding cross validation algorithm to partition a sample dataset and then train and test a set of N classifiers with the dataset folds. At each test iteration, trained models of the set of classifiers generate probabilities that a sample has a flaw, resulting in a set of N probabilities or predictions for each sample in the test data. With a sample size of S, the system passes the S sets of N predictions to a logistic regressor along with 'ground truth' for the sample dataset to train a logistic regression model. The trained classifiers and the logistic regression model are stored as the classifier ensemble.

Open source vulnerability prediction with machine learning ensemble

To analyze open-source code at a large scale, a security domain graph language ('SGL') has been created that functions as a vulnerability description language and facilitates program analysis queries. The SGL facilitates building and maintaining a graph database to catalogue vulnerabilities found in open-source components. This graphical database can be accessed via a database interface directly or accessed by an agent that interacts with the database interface. To build the graph database, a database interface processes an open-source component and creates graph structures which represent relationships present in the open-source component. The database interface transforms a vulnerability description into a canonical form based on a schema for the graph database and updates the database based on a determination of whether the vulnerability is a duplicate. This ensures quality and consistency of the vulnerability dataset maintained in the graph database.

Growing community of inventors

Singapore, Singapore

Asankhaya Sharma