Rostyslav Bodnar

Data Analyst

rossbodnar@protonmail.com

Projects

Lost and Stolen Weapons
in Ukraine ❯

An insightful dashboard of data on lost and stolen weapons in Ukraine. By utilizing the dataset released by the Ministry of Internal Affairs, it highlights trends and key metrics through interactive charts and graphs, making the information accessible and understandable to the public.

Obtained the dataset from data.gov.ua while ensuring compliance with legal and ethical standards.
Examined the dataset structure and variable types to understand the data's characteristics.
Conducted comprehensive data cleaning, including handling missing values, removing duplicates, and resolving inconsistencies.
Applied feature engineering techniques to create new features or transform existing ones, enhancing the dataset's utility for analysis.
Exported the cleaned and enriched dataset to a suitable format for subsequent Exploratory Data Analysis (EDA).

Identified key metrics and variables of interest based on the dataset's context and objectives.
Utilized statistical techniques and aggregations to summarize and explore the data, revealing insights and patterns.
Created visualizations, such as histograms, scatter plots, and bar charts, to explore the data's distribution, correlations, and trends.

Designed layout and composition of the Streamlit page.
Selected metrics and visualizations from EDA.
Populated the page with placeholders for content and visualizations.

Refactored Jupyter notebook code into separate Python files.
Developed scripts for automated data processing, modeling, and visualization.
Created a cron job configuration file for scheduled script execution using GitHub Actions.
Set up GitHub Actions for continuous integration.

Created a data flow diagram for maintenance facilitation.
Refined the GitHub repository.
Deployed the project to Streamlit cloud hosting.

Added new annotations with important events to the main bar chart.
Monitored GitHub Actions to ensure successful workflow completions.
Reviewed pipeline scripts logs to identify new weapons in the dataset.

Find Military Gear UA Bot ❯

Telegram bot helping users to search for up to date and in-stock military clothing and gear prices across 21 ukrainian online stores.

Developed a basic scraper for initial data extraction.
Compiled a list of websites and relevant details in an Excel sheet.
Created a WebScraper class to encapsulate scraping logic.
Implemented custom RegEx for easier addition of new stores.
Included duplicate content detection.
Enabled asynchronous scraping for improved performance.
Developed a CLI for the scraper with additional functionality (separate repo).

Created a simple initial version of the Telegram bot.
Integrated the bot with the scraping script.
Ensured functionality for multiple concurrent users.
Enhanced UX/UI for straightforward interaction.
Added basic usage data tracking.

Improved code readability.
Addressed performance bottlenecks.
Implemented logging for better error handling.
Added user input validation.
Enhanced query and product name matching.

Created and published a Docker image to a private repository on Docker Hub.
Deployed the bot image on Oracle Cloud.
Refined the GitHub repository.

Monitored logs to ensure continued support for current stores.
Looked for new online stores to add.

Network Visualization of League of Legends Champions ❯

Interactive network visualization that connects champions from the game League of Legends to their respective classes. Each champion's node is color-coded to reflect their difficulty level, providing an immediate visual indication of the champion's complexity. Tooltips offer additional information about each champion ratings.