Create Your Own Music Recommendation System Using Machine Learning
Half-day workshop - in English
Have you ever wondered how Youtube and Spotify can possibly recommend new videos and songs to you? Having heard that machine learning plays an important role, you may be wondering what all the fuss is about. How does it work? How can you get started using it yourself?
This session serves as an introduction to machine learning, based on a practical use case. We will be utilizing Spotify's API to extract features from music, before we visualize and cluster the data, and also train classifiers for discovering new music. We will give you an introduction to several algorithms used for clustering and classification of data. In addition to digging into some traditional machine learning algorithms, such as K-means and SVM, we will also take a look at artificial neural networks, which in recent years have produced remarkable results in various fields. For all of this, we will be using frameworks like Keras and Sklearn.
If machine learning has been a mysterious domain to you, this session will most likely leave you with a greater understanding of the process and aid you in how to set up projects of your own.
Primarily for: Developers
Participant requirements: Bring your laptop and an eagerness to learn ;) Code repo is available through git: https://github.com/knowit/ml-music-rec. Recommended IDE: PyCharm by Jetbrains. If possible, we would like participants to set up their python environment and project repo ahead of the workshop. Thus, we can spend most of the workshop time helping you solve the tasks, rather than technical issues :) We will be programming in Python 3.6 and will be using a number of python packages, which are listed in the requirements.txt file. By doing step 1 and 2 in the readme of the repo, you're good to go for the workshop. If you run into difficulties, do not hesitate to shoot us an email: email@example.com firstname.lastname@example.org email@example.com Note: All three of use macOS daily. If you intend to use Windows and run into trouble, we will of course try to help you out the best we can, but if you have the luxury of using a Unix-based system (macOS, Linux), that would definitely be preferable. Of course, if you are confident with setting up a python environment on Windows, go for it :)