Data Science Portfolio

Repository containing portfolio of data science projects completed by me for academic, self learning and hobby purposes.

Contents

Machine Learning

Predicting Car Price

A model to predict the value of a given car on a car sales portal: otomoto

Tools: scikit-learn, pandas, numpy, seaborn, matplotlib
Dataset: (self-made: Web scraping) https://drive.google.com/file/d/1xp0Aam0Vv-A3zTQ-kihz0wExzqdpJjeZ/view

Data Analysis and Visualization

Otomoto Dataset - Price Analysis

Analysis of factors influencing the selling price of a car on otomoto

Tools: pandas, numpy, seaborn, matplotlib
Dataset: (self-made: Web scraping) https://drive.google.com/file/d/1xp0Aam0Vv-A3zTQ-kihz0wExzqdpJjeZ/view

TFT Masters France Tournament Summary

Summary of TFT games at the Masters France tournament. Includes composition visualization and statistics of units, synergies and items.

Tools: Google Sheets, Excel, Python
Dataset: Self-collected data from Riot API

TFT Reckoning Worlds HP per Round Visualization

Visualization of the change in health level of TFT World Championship players per game.

Tools: Google Sheets, Excel, Python(OpenCV, Tesseract OCR), Azure OCR
Dataset: Data was collected automatically from VOD using Azure OCR and Tesseract OCR

SCH Twitter Profile

Twitter profile publishing stats from the Teamfight Tactics tournament scene.

Tools: Google Sheets, Excel, Python(OpenCV, Tesseract OCR), Azure OCR
Dataset: Self-collected data from Riot API and VOD (using computer vision)

NLP

Sentiment Analysis Models Comparison

Implementation and comparison of models for opinion sentiment classification in Polish.

Tools: pandas, numpy, matplotlib, scikit-learn, tensorflow
Dataset: https://drive.google.com/file/d/1vXqUEBjUHGGy3vV2dA7LlvBjjZlQnl0D/view?usp=sharing