Provides interactive visualizations for text classification using the Naive Bayes Classifier​. It aims to demystify the inner workings of the algorithm, enabling users to gain deeper understanding of how it categorizes text documents. It serves as a tool for identifying instances when the algorithm's performance falls short. Frequently, it's quite challenging to pinpoint the reasons behind incorrect classifications, and the decision-making process driving predictions often remains obscured. It offers a window into the algorithm's decision-making process, illuminating these aspects and making them more transparent.

🌐 Check the app:

• the app is hosted here: https://freezpmark-dashapp.odooserver.sk/
• the app within my website with some explanation: https://freezpmark.odooserver.sk/projects-machine-learning

🧑‍💻 Running locally:

• git clone https://github.com/freezpmark/dash-app-naive-bayes-visualization
• cd dash-app-naive-bayes-visualization
• pip install -r requirements.txt
• python dashWEB_eng.py

Interactive tutorials in Machine Learning — Bachelor’s thesis (Nov 2017 - May 2018)

• Created Dash web app with interactive visuals of Naive Bayes classifier, enhancing its interpretability and text analysis options
• Pioneered an enhanced learning approach beyond standard Jupyter Notebook tutorials, resulting in an 18% improvement in the learning experience, evidenced by 28 learning participants' quiz performance
• Earned an A+ grade in data visualization and a B grade for thesis that focuses on Natural Language Processing

[Python, NumPy, Plotly Dash]