Clustering na machine learning task wey dey try find objects wey resemble each oda and group dem into groups wey dem dey call clusters. Wetin make clustering different from oda methods for machine learning be say e dey happen automatically. In fact, you fit talk say e be opposite of supervised learning.
Nigeria get plenty different people and dem music taste dey different too. Using data wey dem collect from Spotify (inspired by dis article), make we check some music wey dey popular for Nigeria. Dis dataset get data about different songs like 'danceability' score, 'acousticness', loudness, 'speechiness', popularity, and energy. E go dey interesting to see wetin we fit discover for dis data!
Photo by Marcela Laskoski on Unsplash
For dis series of lessons, you go learn new ways to analyze data using clustering techniques. Clustering dey very useful when your dataset no get labels. If e get labels, classification techniques like the ones wey you don learn for previous lessons go make more sense. But if you wan group data wey no get labels, clustering na better way to find patterns.
Some low-code tools dey wey fit help you learn how to work with clustering models. Try Azure ML for dis task
Dis lessons na Jen Looper write am with 🎶, and Rishit Dagli plus Muhammad Sakib Khan Inan help review am.
The Nigerian Songs dataset na from Kaggle dem collect am as dem scrape am from Spotify.
Some useful K-Means examples wey help create dis lesson include dis iris exploration, dis introductory notebook, and dis hypothetical NGO example.
Disclaimer:
Dis dokyument don translate wit AI translation service Co-op Translator. Even though we dey try make am accurate, abeg sabi say machine translation fit get mistake or no dey correct well. Di original dokyument for im native language na di main source wey you go trust. For important information, e better make professional human translator check am. We no go fit take blame for any misunderstanding or wrong interpretation wey fit happen because you use dis translation.