music data dashboard

data science project

Welcome to my 2020 Spotify music dashboard which was my first project in the world of data science. Developed as a fun and educational venture, this project revolved around constructing an interactive data dashboard using 2020’s top music chart data. The dashboard is accessible at musicdash.streamlit.app. The main focus of this project was the exploration of Natural Language Processing (NLP), a fascinating domain of data science that I’ve delved into to decipher patterns and trends in popular music. Through this dashboard, I’ve ventured into basic NLP concepts like text processing, tokenisation, stemming, lemmatisation and have undertaken sentiment analysis on song titles to capture the emotional essence of popular tracks. Additionally, this project has been an opportunity to experiment with text classification techniques, aiming to categorize songs by their genre or mood. This project was powered by Python and the user-friendly Streamlit framework. I’m open to ideas, feedback, and collaboration from fellow enthusiasts. Stay tuned for more updates as this project continues to evolve!

Title and genre chart, as well as the button to take user to the original data used.

Delving into sentiment analysis has been a truly fascinating experience, uncovering the subtle emotional undercurrents hidden within the simple titles of songs/

Sentiment analysis chart.

Finding insights in the data based on music popularity and its correlation with musical characteristics was fascinating and a rich learning experience. Some of the insights found can be seen below.

Chart to find patterns in most common songs in terms of musical characteristics

To further explore the world of NLP, I learned about concepts such as association rule mining and N-gram analysis, which I applied to playlist titles in the dataset. I also learned about the common use of word clouds, which I applied to song names and artist titles.

Wordclouds generated and association rule mining results

Through this project, I gained a strong understanding of NLP techniques, and learned concepts like text processing, tokenisation, and sentiment analysis to reveal the emotional and thematic trends in music. Additionally, I’ve honed my skills in data visualisation and interactive dashboard creation using Python and Streamlit, effectively translating complex data insights into user-friendly formats.