Data Science Portfolio

Welcome to My Data Science Portfolio

This portfolio contains projects completed by me for academic, self-learning, and hobby purposes. All projects are hosted on GitHub. For an in-depth view of my profile, please visit my GitHub profile.

Direct links to my completed projects can be found below.

ABOUT ME

I’m a data science enthusiast with a passion for extracting insights and solving problems through data analysis. My journey began with engineering studies at the University of South Africa, where I developed a strong foundation in programming principles. However, my true calling emerged when I discovered the exciting world of data science. Fueled by this passion, I honed my skills and excelled in this field, leveraging my programming background to its full potential.

My commitment to data quality and best practices is reflected in my certification as a Data Management Professional (DAMA). Additionally, I hold a certification in Microsoft T-SQL programming, demonstrating expertise in a valuable database language.

Data-Driven Accomplishments

As a data scientist, I thrive on tackling real-world challenges through innovative solutions. Here are some highlights of my professional journey:

Spearheaded Data Migrations: I’ve led complex data migration initiatives, meticulously extracting and transforming critical information for enterprise-level cloud adoption. My deep understanding of the ETL (Extract, Transform, Load) process ensures data integrity and accessibility throughout the migration. Forecasting Expertise: I possess expertise in building robust forecasting models, including those leveraging ARIMA techniques for time series data analysis. This empowers data-driven decision making within organizations. GIS Analysis Pro: I actively contribute to GIS analysis projects, utilizing spatial data to uncover valuable insights and inform strategic planning. Machine Learning Implementations: I’ve constructed machine learning models for various tasks, including time series forecasting. This demonstrates my ability to apply cutting-edge algorithms to solve real-world problems. Data Architecture Prowess: My participation in data architecture projects has solidified my understanding of data infrastructure design and implementation. Continuous Learning & Innovation

I’m a firm believer in staying ahead of the curve. I actively research and experiment with the latest advancements in AI and Machine Learning. To keep my skills sharp, I build and iterate on mini models, furthering my understanding of these powerful technologies.

Beyond Data

Beyond data, I enjoy a good challenge, both online and over the chessboard. When I’m not immersed in the world of data, you might find me exploring various activities and seeking new experiences.

Certifications:

Projects in progress 2024:

AI Communication Coach (in progress)

This project aims to develop a platform that utilizes Artificial Intelligence (AI) to assist users in improving their communication skills for public speaking and interviews.

Current Functionality:

Speech Recognition: Capture and analyze user speech
Text Analysis: Perform sentiment analysis, keyword extraction, and named entity recognition
LLM Integration: Use pre-trained LLMs to generate responses based on user input

Video showcasing the current features

Future Development:

Enhance response generation using the LLM with context awareness and conversation history
Integrate feedback mechanisms to improve the coach’s responses over time
Explore additional features like voice synthesis for coach responses or sentiment visualization

Tools used:

Python: Programming Language
Jupyter Notebook: Development environment
SpeechRecognition: Speech Recognition
NLP Libraries: NLTK, spaCy, TextBlob
Transformers: Hugging Face Transformers for text generation
(Optional) OpenCV: Computer Vision for future features

Mini Project - CODING IN PROGRESS

Renewable Energy Usage Prediction:

Using machine learning models to predict trends in renewable energy usage based on historical data and various socio-economic factors.

Tools used:

Python
Pandas
NumPy
Scikit-learn
Matplotlib
Seaborn

Projects

Final College Data Science Project:

This project/report covers:

Statistical analysis, PowerBI reporting and app deployment, Python DML. All scenarios, code, and data source information are included in the report.

)

Tools used:

PowerBI
Python
SQL
Jupyter Notebook
Statistical methods

Personal Projects:

Loan Dataset - Exploratory Data Analysis:

Exploratory analysis of the Prosper Loans company loan data using Pandas and Seaborn visualizations.

Tools used:

Pandas
Seaborn
Matplotlib

Credit Fraud Detection: Prediction/Analysis:

Model that predicts transactional credit fraud by utilizing machine learning techniques.

Tools used:

Pandas
NumPy
NLTK
Sklearn
Seaborn
Matplotlib

Stock Market Analysis for Tech Stocks:

Analysis of technology stocks, including change in price over time, daily returns, and stock behavior prediction.

Tools used:

Pandas
Folium
Seaborn
Matplotlib

Titanic Dataset - Exploratory Analysis:

Exploratory analysis of the passengers onboard RMS Titanic using Pandas and Seaborn visualizations.

Tools used:

Pandas
Folium
Seaborn
Matplotlib

Unsupervised Learning: Creating Customer Segments:

Analyzing a dataset containing data on various customers’ annual spending amounts (reported in monetary units) of diverse product categories for discovering internal structure, patterns, and knowledge.

Tools used:

Scikit-learn
Pandas
Seaborn
Matplotlib
Pygame

Simple COVID-19 Self Screening Program using Python:

The program will tell if someone has COVID-19 or not, based on a number of preset questions.

NOTE: This project is purely to demonstrate that not all problems require complex solutions. Feel free to test it out.

Tools used:

Python
github codespace

Jermaine Varnicker