Master of Science in Cyber Security (M.S.C.S.), Wright State University, 2023, Computer Science

Malware detection is a critical task in ensuring the security of computer systems. Due to a surge in malware and the malware program sophistication, machine learning methods have been developed to perform such a task with great success. To further learn structural semantics, Graph Neural Networks abbreviated as GNNs have emerged as a recent practice for malware detection by modeling the relationships between various components of a program as a graph, which deliver promising detection performance improvement. However, this line of research attends to individual programs while overlooking program interactions; also, these GNNs tend to perform feature aggregation from neighbors without considering any label information and significantly suffer from over-smoothing on node presentations. To address these issues, this thesis constructs a graph over program collection to capture the program relations and designs two enhanced graph convolutional network (GCN)architectures for malware detection. More specifically, the first proposed GCN model in-corporates label propagation into GCN to take advantage of label information for facilitating neighborhood aggregation, which is used to propagate labels from the labeled nodes to the unlabeled nodes; the second proposed GCN model introduces residual connections between the original node features and the node representations produced by GCN layer to enhance the flow of information through the network and address over-smoothing is-sue. The experimental results after substantial experiments show that the proposed models outperform the baseline GCN and classic machine learning methods for malware detection, which confirm their effectiveness in program representation learning using either label propagation or residual connections and malware detection using program graph. Furthermore, these models can be used for other graph-based tasks other than malware detection, demonstrating their versatility and promise.

Committee: Lingwei Chen Ph.D. (Advisor); Meilin Liu Ph.D. (Committee Member); Junjie Zhang Ph.D. (Committee Member) Subjects: Computer Science; Information Science

PhD, University of Cincinnati, 2022, Engineering and Applied Science: Computer Science and Engineering

Identification of the mechanism of action (MoA) of a small molecule which causes pharmacological effects on cellular networks governing gene expression levels is an important field of study for the purpose of drug development and repurposing. While gene expression can be used for the prediction of small molecule MoA using traditional machine learning algorithms, these algorithms do not consider the underlying complexity of cellular level biological networks driving gene expression. In particular, capturing predictive features from the polarity of interaction in cell signaling networks where nodes in the network either activate or inhibit other nodes is still a challenging problem for the prediction of drug MoA. We propose an ensemble deep learning meta-algorithm for predicting small molecule MoA from gene expression data using unsigned and signed graph convolutional networks (GCN). We developed a GCN algorithm to extract features from signed networks and combined predictive probabilities with that of an unsigned GCN using stacking. Our ensemble methodology improves the overall predictive capabilities significantly when compared to unsigned or signed GCN.

Committee: Mario Medvedovic Ph.D. (Committee Member); Gowtham Atluri Ph.D. (Committee Member); Ali Minai Ph.D. (Committee Member); Jaroslaw Meller Ph.D. (Committee Member); Raj Bhatnagar Ph.D. (Committee Member) Subjects: Bioinformatics