Traffic Forecasting With Vset-Nets: A Vgae Spatial Embedding For Temporal Networks Approach,
2025
San Jose State University
Traffic Forecasting With Vset-Nets: A Vgae Spatial Embedding For Temporal Networks Approach, Mrunmayee Dhapre
Master's Projects
Traffic forecasting is important for improving transportation systems by enabling better traffic management, congestion reduction, and urban planning. However, predicting traffic accurately is challenging due to the strong spatial dependencies between different road segments and the temporal changes in traffic patterns over time. Traditional time-series and graph models often struggle to capture both of these aspects effectively. In response, recent research has focused on temporal graph representation learning methods that jointly consider spatial relationships and temporal features in networks. This project proposes a hybrid model called VSET-Nets (VGAE Spatial Embedding for Temporal Networks) that employs Variational Graph Autoencoders (VGAEs) for …
Domain-Specific Graph Rag Pipelines: Optimized Approaches For Building Efficient Personal Knowledge Repositories,
2025
San Jose State University
Domain-Specific Graph Rag Pipelines: Optimized Approaches For Building Efficient Personal Knowledge Repositories, Omkar Yadav
Master's Projects
Managing personal data, including notes, calendar events, to-do lists, and other personal information, has become increasingly complex and challenging. In response to this issue, I propose a framework using RAG that enables a large language model (LLM) to efficiently query this data without requiring training on the personal data itself. Conventional retrieval systems, including those leveraging vector-based retrievalaugmented generation (RAG), are effective at handling basic queries but struggle to deliver coherent global abstractions, integrate diverse knowledge sources, and account for temporal nuances. This research explores domain-specific Graph based RAG frameworks that incorporate a knowledge graph to better model relationships, thereby …
Improving Contextual Retrieval For Long Documents In Q & A Systems,
2025
San Jose State University
Improving Contextual Retrieval For Long Documents In Q & A Systems, Sakshi Garg
Master's Projects
With the vast amount of information available on the internet distributed across several lengthy documents, finding relevant information has become more important and challenging. The goal of this project is to develop advanced techniques to retrieve information from long texts in order to deliver accurate and relevant results while ensuring speed and efficiency. As part of this work, we employ techniques to address unique difficulties posed by large and complex documents. This paper presents a custom Retrieval-Augmented Generation (RAG) framework designed to improve contextual retrieval in long and multi-document settings. In this paper, we employ several techniques like summarization, semantic …
Galora: A Lightweight Graph-Aware Llm Framework For Node Classification On Text-Attributed Graphs,
2025
San Jose State University
Galora: A Lightweight Graph-Aware Llm Framework For Node Classification On Text-Attributed Graphs, Mayur Choudhary
Master's Projects
With the exponential rise of language models (LMs) and their potential to understand semantic relationships, large LMs are being used across a wide range of applications. Text-attributed graphs (TAGs) are one notable example where LLMs can be combined with Graph Neural Networks (GNNs) to enhance node classification results. TAGs associate textual content with each node and are commonly seen in various domains such as social networks, citation graphs, recommendation systems, etc. Effectively modeling TAGs would enable deeper insights into different aspects of the graph and improve decision-making in relevant domains. We present GaLoRA, a parameter-efficient framework to integrate structural information …
Evorgcn: Harnessing Esm-2 Evolutionary Embeddings With Relational Gcns For High-Fidelity Protein-Protein Interaction Prediction,
2025
San Jose State University
Evorgcn: Harnessing Esm-2 Evolutionary Embeddings With Relational Gcns For High-Fidelity Protein-Protein Interaction Prediction, Mohit Kunder
Master's Projects
Accurately predicting protein-protein interactions (PPIs) is essential for understanding cellular function and advancing biomedical discovery. We model PPIs as graphs, where nodes represent proteins and edges denote interactions. Using interaction data from the STRING database, we use two samples of it, namely the benchmark datasets—SH27K and SH148K—filtered by confidence score and annotated by interaction mode (multiple relations). In this project, we present EvoRGCN, a graph-based machine learning framework for PPI prediction that integrates both sequence-level (ESM-2 embeddings) and network-level information. We incorporate various Graph Neural Network architectures, including Graph Convolutional Networks, Graph Attention Networks, and Relational Graph Convolutional Networks. Our …
Link Failure Localization In Hierarchical, Multidomain Optical Networks,
2025
San Jose State University
Link Failure Localization In Hierarchical, Multidomain Optical Networks, Martin Mihailov Bojinov
Master's Projects
Optical networks transport data encoded on light signals over optical fiber cables. Individual optical networks are managed by domain administrators, such as service providers, vendors, or regional bodies. Multidomain optical networking explores the possibility of enabling seamless data transmission across domain boundaries. In this project, we explore how we can perform network monitoring in a hierarchical multidomain optical network while preserving domain security and autonomy. This is done by allocating dedicated monitoring trails across various broker abstractions of our network topologies. We propose three heuristic functions that expedite trail selection. Of the three heuristics, our least monitored algorithm performs the …
Dynamic Pricing For Revenue Maximization In 5g Networks With Elastic Network Slicing And Reinforcement Learning,
2025
San Jose State University
Dynamic Pricing For Revenue Maximization In 5g Networks With Elastic Network Slicing And Reinforcement Learning, Naisheel Shah
Master's Projects
Network slicing is a concept that enables a diverse array of network applications with lots of unique service requirements. For the scope of this research, we delve into elastic network slicing, which has a potential benefit for both the providers and users through cost-effective resource utilization. Dynamic pricing of these slice resources is a method for operators to realize the balance of different types of network slices by implicitly communicating the current network state to slice users. We have designed a custom pricing scheme using Deep Reinforcement Learning for elastic network slices that maximizes the revenue of slice providers while …
Optimizing Analytics Storage Strategies For Search Engines And Wiki Platforms,
2025
San Jose State University
Optimizing Analytics Storage Strategies For Search Engines And Wiki Platforms, Sujith Kakarlapudi
Master's Projects
Modern search engines and wiki platforms generate vast quantities of user inter-
/="/">action data such as page views, edits, clicks, and session events. This data must be stored and aggregated efficiently to enable scalable analytics and responsive querying. Yioop, an open-source search engine framework, serves as our primary case study, processing millions of such events to power its indexing and recommendation features. This report explores a shift from the conventional database storage based model to a log based model, in order to improve scalability and write efficiency. A size-limited, append-only logging facility was provided to log analytics events: the …
Optimizing Integrated Access And Backhaul Topology Using Monte Carlo Tree Search,
2025
San Jose State University
Optimizing Integrated Access And Backhaul Topology Using Monte Carlo Tree Search, Srimanth Reddy Ummenthala
Master's Projects
Integrated Access and Backhaul (IAB) plays a central role in enabling scalability and high-throughput in wireless networks, especially in areas where wired backhaul is not feasible. Optimizing the topology of IAB networks is a complex task, involving trade-offs between path loss, node capacity, and link quality. To address this, this project investigates a Monte Carlo Tree Search (MCTS)-based approach to improve overall network performance by maximizing the capacity of network, considering Signal-to-Noise Ratio (SNR) on wireless links. MCTS provides a guided search mechanism to construct topologies efficiently under the constraints and is evaluated against baseline topologies. Experimental results show that …
Comparative Analysis Of Embedding Techniques With Clustering Algorithms For Malware Opcodes,
2025
San Jose State University
Comparative Analysis Of Embedding Techniques With Clustering Algorithms For Malware Opcodes, Ayush Koul
Master's Projects
Malware detection and classification remain critical challenges in cybersecurity, especially as malicious software becomes increasingly sophisticated and prevalent. While much of the work involving embeddings has traditionally relied on supervised learning approaches, there is significant potential in leveraging unsupervised learning techniques to discern hidden structures in malware data. By employing embedding techniques to convert malware samples into high-dimensional vector representations, we can capture the subtle and complex patterns inherent in malicious code without relying on pre-labeled data. This unsupervised approach helps categorize malware into predefined malware families, greatly aiding in developing cybersecurity solutions. In contrast to traditional supervised models that …
Application Of Advanced Convolutional Neural Network With Robust Hashing On Obfuscated Image Based Malware Dataset,
2025
San Jose State University
Application Of Advanced Convolutional Neural Network With Robust Hashing On Obfuscated Image Based Malware Dataset, Sanket Shekhar Kulkarni
Master's Projects
This project report provides in-depth details on the creation of a malware classification system that makes use of Convolutional Neural Networks (CNNs) that have been strengthened by data set obfuscation and strong hashing. We test many CNN architectures, including MobileNet, ResNet, and DenseNet, using rigorous hashing and obfuscation techniques on datasets. The entire pipeline is described in this research, which ranges from the gathering and preprocessing of data sets to the application of novel hashing techniques that boost overall accuracy in classification and increase resilience against malicious attacks. Parallel to this, we show that dataset obfuscation adds an additional level …
Synthetic Malware Generation Using Generative Ai,
2025
San Jose State University
Synthetic Malware Generation Using Generative Ai, Phanidhar Sai Sravan Chandana
Master's Projects
Malware grows in numbers and complexity, evading conventional signature-and anomaly-based defenses and worsening extreme data sparsity and class imbalance problems for machine learning based detection. Generative models, specifically GANs conditioned on contextual embeddings like BERT have proved effective augmenting training corpora to improve classifier accuracy, but these approaches have largely produced family-specific samples In this paper, we propose a generalized augmentation scheme for generating robust malware embeddings for various families. We begin by extracting opcode sequences from 13 malware families and encoding them into three embedding methods: CountVectorizer, TF-IDF, and BERT’s ‘[CLS]‘ vectors. We therefore train standard GANs and Wasserstein …
Malware Generation And Classification Using Pixelcnn,
2025
San Jose State University
Malware Generation And Classification Using Pixelcnn, Mounika Krishna Teja Karumudi
Master's Projects
Malware poses a serious threat to both data privacy and system security. With the wide variety of malware families and the surge in cyber-attacks, the accurate classification of malware is crucial for building effective detection and prevention systems. In recent years, deep learning (DL) methods in computer vision have shown promise in classifying malware by converting malware files into visual representations and applying DL algorithms to classify the resulting images. Among the different approaches to malware family classification, image-based methods have gained significant interest. This research focuses on leveraging DL techniques for image-based classification of malware. The success of identifying …
Reinforcement Learning-Based End-To-End Monitoring Path Selection In Multi-Domain Optical Networks,
2025
San Jose State University
Reinforcement Learning-Based End-To-End Monitoring Path Selection In Multi-Domain Optical Networks, Soham Choudhury
Master's Projects
This paper presents a novel approach for optimizing network monitoring in optical communication systems using Reinforcement Learning (RL). Assuming a multi-domain architecture with limited domain visibility, we simulate multiple optical connections using an optical communications simulation software, GNPy, obtaining key network metrics to model the system. We developed two RL agents: the first agent selects near-optimal monitoring paths based on network states, and the second agent dynamically adapts its selected paths in response to state changes, such as fiber failures or issues with ROADMs. This adaptive approach allows for continuous improvement of network monitoring, ensuring resilience and efficient fault detection. …
Transforming Chatbot Text: A Sequence-To-Sequence Approach To Human-Like Text Generation,
2025
San Jose State University
Transforming Chatbot Text: A Sequence-To-Sequence Approach To Human-Like Text Generation, Natesh Reddy
Master's Projects
With the advancements in Large Language Models (LLMs) such as ChatGPT, the boundary between text written by a human and that created by AI has become blurred. This poses a threat to systems designed to detect AI-generated content. In this work, we adopt a novel strategy to adversarially transform GPT-generated text using sequence-to-sequence (Seq2Seq) models, to make the text more human-like. We concentrate on improving GPT-generated sentences by including significant linguistic, structural, and semantic components that are typical of human-authored text. The goal is to then use this transformed data to train a more robust detection model. Our experiments show …
Enhancing Robustness Of Cnn Model For Malware Detection Using Gan-Based Data Augmentation And Transfer Learning,
2025
San Jose State University
Enhancing Robustness Of Cnn Model For Malware Detection Using Gan-Based Data Augmentation And Transfer Learning, Milind Anand Pathak
Master's Projects
Malware classification is a critical component in the field of cybersecurity. Accurate identification of a malware family can enable timely threat detection and response. In this thesis, we propose a robust image-based malware classification pipeline using Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs), with a focus on improving performance for underrepresented malware families. We train a baseline CNN model on the Malimg dataset across 25 malware families, but observe misclassifications in classes with limited data and overlapping visual features. To address this, we apply targeted augmentations and generate class-specific synthetic data using StyleGAN2-ADA. A CNN trained on the …
Comparative Analysis Of Adversarial Permeability In Cpu-Native, Qat And Onnx-Based Quantized Transformer Models,
2025
San Jose State University
Comparative Analysis Of Adversarial Permeability In Cpu-Native, Qat And Onnx-Based Quantized Transformer Models, Fahad Siddiqui
Master's Projects
This work explores securing and optimization of Transformer-based time series forecasting models. We employ several quantization techniques, including quantization-aware training (QAT), and tested the robustness of quantized models by adversarially attacking them. The preliminary results of this work, in our controlled setup, indicate that quantized models outperform the full precision model in terms of robustness against adversarial attacks. They achieved this robustness while showing a very minimal decrease in their forecasting performance.
Enhancing Code Review Automation With Large Language Models Using Qlora Fine-Tuning And Rags,
2025
San Jose State University
Enhancing Code Review Automation With Large Language Models Using Qlora Fine-Tuning And Rags, Sumukh Naveen Aradhya
Master's Projects
In this technological era where Artificial Intelligence and Machine Learning are revolutionizing various domains, Large Language Models (LLMs) are emerging as a very powerful tool. In the software development lifecycle, it is extremely important to have reliable code reviews to ensure security and maintain code quality. This project aims to survey various existing methodologies to aid creation of efficient code review automation agents and also research on ways to make this process more efficient. Parameter Efficient Fine-Tuning (PEFT) methodologies such as LoRA and QLoRA have been explored with an additional focus on a hybrid model that combines adaptive QLoRA with …
Multilingual Sentiment Analysis Using Ensemble Learning,
2025
San Jose State University
Multilingual Sentiment Analysis Using Ensemble Learning, Farhan Ansari
Master's Projects
The widespread use of multiple social media platforms has amplified the expression of public opinions over the Internet in languages such as English, Hindi and Spanish. With the aid of technological advancements in machine learning, we can analyze opinions posted on the Internet and gauge public sentiments. There are organizations and businesses that are interested in the evaluation of these sentiments as these type of data can generally be used to obtain the opinion of a product, restaurant, a candidate, etc. In this study, we perform a comparative analysis of three popular ensemble learning methodologies (Boosting, Bagging and Stacking) based …
Clustering Organ Cell Types,
2025
San Jose State University
Clustering Organ Cell Types, Venkata Satya Swathi Mattaparthi
Master's Projects
The Human Cell Atlas (HCA) created a reference map of all human cells. My project uses the Tabula Sapiens dataset, developed under HCA and based on single-cell RNA sequencing data, to explore cell type and tissue diversity. I performed experiments using the Elbow method and a formula based on dataset observations to determine the number of clusters, then applied k-means clustering on two representative subsets of the All Cells dataset. Clusters were selected for analysis using Shannon’s Diversity Index and Pielou’s Evenness. A novel algorithm based on the cell differentiation tree was used to validate the biological coherence of the …
