Shu Wang's Projects
Implementation of Adaboost with decision stump classifiers.
Implementation of Adaboost with Perceptron Classifiers.
Open Source Code for Machine Learning in Computer Vision
Provenance Analysis of Binary Code with RNN model.
Data Analysis for Human Trafficking
Inference from Fingerprint Images
Faster R-CNN for Salient Object Detection.
2016 Code Craft Challenge
A Implementation of HMAC Encryption by using SHA-1 Hash Function
K Nearest Neighbor and Condensed KNN Algorithm
Mobile Phone Detection with Modulated Replay Attack
Perceptron Learning Algorithm and Linear Regression
Basic Machine Learning Classifiers with scikit-learn.
The source code in the paper "When the Differences in Frequency Domain are Compensated: Understanding and Defeating Modulated Replay Attacks on Automatic Speech Recognition". This paper is published in the ACM Conference on Computer and Communications Security (CCS) 2020.
Naive Bayes and Logistic Regression with Natural Language Processing
Real or Not? NLP with Disaster Tweets.
Feed Forward Neural Network for Sentiment Classification and Language Modeling
Recurrent Neural Network for Natural Language Processing
Semantic Role Labeler in Natural Language Processing
Paimon: Patch Identification Monster (extended version of GraphSPD)
Data clearance for security patches and non-security patches. This method is described as Nearest Link Search in the paper "PatchDB: A Large-Scale Security Patch Dataset", which appears in 2021 51st Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN 2021), Taipei, June 21-24, 2021, pp. 149-160.
A Demo Program of Security Patch Identification with Graph Neural Networks.
Security Patch Identification with Classical Machine Learning Algorithms.
Oversampling operations on security/non-security patches. This method is described in the paper "PatchDB: A Large-Scale Security Patch Dataset", which appears in 2021 51st Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN 2021), Taipei, June 21-24, 2021, pp. 149-160.
Patch oversampling (synthesis) with direct patch analysis. This is an alternative solution to the PatchOversampling repository, providing a simpler and more direct way to synthesize patches. The original oversampling method is described in the DSN'21 paper "PatchDB: A Large-Scale Security Patch Dataset".
The source code in the paper "PatchRNN: A Deep Learning-Based System for Security Patch Identification". This paper appears in the 2021 IEEE/AFCEA Military Communications Conference (MILCOM 2021), San Diego, USA, November 29–December 2, 2021.
A demo program of security patch identification using the RNN model, which is demonstrated in the paper "PatchRNN: A Deep Learning-Based System for Security Patch Identification". This paper appears in the 2021 IEEE/AFCEA Military Communications Conference (MILCOM 2021), San Diego, USA, November 29–December 2, 2021.
RNN-based security patch identification with oversampling samples. This is an extension code in the MILCOM'21 paper "PatchRNN: A Deep Learning-Based System for Security Patch Identification".
Logistic Regression Implementation using Python
The source code in the paper "Secure In-Vehicle Automatic Speech Recognition Systems". This paper is published in the 23rd International Symposium on Research in Attacks, Intrusions and Defenses (RAID) 2020.