rafailskoulos17 Goto Github PK

complex_networks

cyber_data_analytics

Implementations for the lab assignments of Cyber Data Analytics (CS4035), a MSc course in TU Delft. The topics that we worked on are credit card fraud detection, anomaly detection in SCADA networks and streams and malware detection.

data_visualization

Developed two visualization projects with two fellow classmates. The first project is related to Information Visualization (InfoVis), where and we chose our own topic (NBA statistics) and conducted exploratory visual analysis. The second project is about Volume Visualization (VolVis) and we developed a volume renderer based on the raycasting approach. For the former, we used the D3 library of JavaScript, while the latter was fully implemented in Java.

handwritten_digits_classification

Implemented in Matlab, in a group of 3, a Handwritten Digits identification system by exploiting several machine learning techniques.

ir_core_project

malicious_php_detection

Build a system which is able predict if a given php file is malicious or benign, by using machine learning methods. To achieve this I extracted several lexical features and we chose the best among of them. Also I experimented with several different classifiers found in literature, namely Support-Vector Machine(SVM), Logistic Regression and Decision Tree.

netflow_classification

netflow_embeddings

Experimenting with applying word embeddings in netflow data

supercomputing-for-big-data

Developed a big data pipeline for identifying important events from the GDELT Global Knowledge Graph, as a series of 3 assignments. In the first two assignments, a Spark application for processing the GDELT dataset was written in Scala and deployed in a big data cluster on AWS. In the final assignment, the application was modified to work in a streaming data context using Apache Kafka.

temporalopponentmodeling

web_data_management_group_9

Implementing Microservices with different database Backends: Cassandra vs. Postgres