A distributed implementation of "Nested Subtree Hash Kernels for Large-Scale Graph Classification Over Streams" (ICDM 2012).

The goal of this project is aimed at optimizing Bank Marketing Model through building an event streaming pipeline around Apache Kafka and its ecosystem that communicates with a Machine learning model microservice. Utilizing this to display the likelihood and status of Bank Customers in real time.

Online Anomaly Detection

Streaming data analysis using AWS tools such as Cloud9 to generate events in the cloud, using boto3 to send records to Kinesis Data Firehose to connect to the S3 bucket destination, saving files in .parquet format. With the help of Glue, a data catalog will be created to enable real-time querying of all records with Athena.

Apache Spark With Scala - hands on with big data

Netflow sample

Python code is shared that simulates random events in two scenarios: Technology E-commerce and Megastore in their mobile app. This is done to generate large-scale data that can be processed using Data Engineering tools.

High-frequency arbitrage detection system built with Python and Apache Flink. Features a decoupled microservices architecture with Kafka streaming, real-time windowed analytics, a Streamlit executive dashboard, and automated Discord notifications.