BACEM BEN DALY

Elon Musk's tweets influence on cryptocurrencies prices

· Scraped tweets from Elon Musk's Twitter account using Python web scraping library Scrapy.

· Used the Roberta sentiment analysis model to classify each tweet as positive, negative, or neutral.

· Retrieved historical cryptocurrency prices for relevant coins and merged it with the sentiment data.

· Created interactive visualizations in Tableau to present the results.

Techniques : Scrapy – Roberta - Tableau 

Research advisor: Dr. Jamie Humphries - humphriejk@sfasu.edu

Octopus Predictions

· Wrote web scraping scripts using BeautifulSoup to extract historical sports data from various websites.

· Scheduled and automated web scraping processes to ensure up-to-date data

· Preprocessed and cleaned scraped data to handle missing values, outliers, and inconsistencies

· Employed scikit-learn for feature engineering, creating predictive models, to forecast future games outcomes

· Accomplished cross-validation and hyperparameter tuning to improve model accuracy

· Created a user-friendly web application using Angular for visualizing historical sports data and game predictions

Techniques: BeautifulSoup - Python scikit-learn - NestJS - MongoDB - Angular 

Health Metrics Monitoring Application

· Built data ingestion processes using Kafka to collect real-time data streams from various sources

· Set up PySpark jobs to process and transform raw streaming large volumes of data efficiently.

· Created Docker-compose configurations to containerize the different components of the data pipeline.

· Utilized Python Bokeh to create interactive and real-time dashboards for visualizing streaming data.

· Integrated data from Kafka, PySpark, and MongoDB into the dashboard.

Techniques: Docker-compose - Hadoop - Kafka - PySpark - MongoDB - Python Bokeh

Mini Java Compiler

· Implemented a lexical analyzer using Flex to tokenize Java source code

· Created a parser using Bison to perform syntax analysis

· Coded a semantic analyzer to perform type checking and ensure semantic correctness of the Java code

· Generated intermediate representation (IR) code from the parsed Java source code

Techniques : C++ - Flex - Bison