Machine learning uses algorithms to learn patterns and regularities in data. Such statistical models are suitable for predicting events. The participants practically work out the creation of a model for regression analysis using a continuous data example (Spotify data). Regression analysis is one of the statistical analysis techniques. It models relationships between a dependent variable and one or more independent variables. The online service Spotify stores audio properties, e.g., energy and volume, for musical works. In the course, participants will use the model created to examine pieces of music and determine properties predictively.
Titel: Machine Learning – Predicting properties on Spotify data
Next Session: 06.09.2022, 3 p.m. – 6 p.m. (Speakers: Anja Neumann, Timo Adameit, Jan Ewald, Thomas Burghardt)
Target group: Beginner on Python, basic knowledge on Pandas
Format: Tutorial, hybrid
- Introduction to general aspects of machine learning (10%).
- Tutorial on creating a model to predict values on Spotify data (90%)
The following handouts (slides, example applications) will be provided to the participants:
- PDF on “Introduction to Machine Learning”
- CSV file (Spotify data)
- Jupyter notebook for working with Python, Pandas and sklearn
- Basic knowledge of Python 3.x (desirable)
- Basic knowledge of Python libraries Pandas, Numpy and sklearn (desirable)
- Basic knowledge in using Jupyter notebook
After the training, participants will know and be able to use the regression model selected by the trainees to predict values.
Check out the other trainings by ScaDS.AI Dresden/Leipzig.