Analysis performed on data from the Steam platform using Apache Spark and Cloud services such as Amazon Web Services.

TU Berlin Cloud Computing - correctly implemented assignment4

Analysis of Airline On Time Performance Dataset

Define a big data architecture and perform distributed machine learning calculations on an EMR cluster using AWS

Realtime data pipeline

A Cloud based Reddit stock sentiment analyzer that analyzes overall sentiment from a configurable selection of stock subreddits for each stock. The architecture utilizes AWS MSK (Kafka), AWS EMR (PySpark) and AWS Lambda (Python 3) for maximum scalability and the OpenAI API for sentiment analysis through prompt engineering.

ETL Pipeline extracts JSON files from AWS S3 bucket and transforms these using an AWS EMR Spark Cluster and stores the data into an AWS S3 bucket in parquet file format.

A scalable prototype of an image recognition engine deployed on AWS.

MLP for Sentiment Analysis on Movie's Reviews.

Built a data model, data warehouse and pipeline for extracting transforming and loading data into a star schema-based data model in a redshift database

In this repo, I build a LogisticRegression prediction model with Dask and PySpark and initialize an AWS EMR cluster to run the entire pipeline.

A CNN is deployed in AWS to extract image features in the context of distributed computing.

Implemented random forest machine learning algorithm using pyspark on AWS EMR to classify the wines. The model is then deployed in docker container.

Udacity project: implementing an ETL to process data with Apache Spark and store them in AWS S3 storage

Predicting customer churn for the music app, Sparkify, using PySpark on AWS EMR clusters

An opinionated framework for running big data jobs

Data Pipeline Analytics Platform is an end-to-end generic Big Data pipeline. Involves following tech stack: AWS S3, AWS Redshift, AWS EMR Cluster, Apache Spark, Apache Airflow.