Hands-on training about data clustering with orange data mining toolbox.

Area of Science:

• Data Science
• Educational Technology

Background:

• Data clustering is a fundamental data science technique.
• Existing training often requires advanced statistical or computational backgrounds.
• Intuitive algorithms and interpretable results make clustering ideal for introductory data science.

Purpose of the Study:

• To propose a novel, hands-on training methodology for data clustering.
• To make data clustering accessible to a general audience without prerequisites.
• To enhance engagement through practical application and visual exploration.

Main Methods:

• Problem-based learning starting with raw data.
• Gradual introduction of data processing and analysis techniques.
• Emphasis on visual representations of data and models.
• Exploratory data analysis with experimentation.

Main Results:

• A structured pedagogical approach for data clustering education.
• Detailed curriculum including datasets and analysis workflows.
• Demonstration of a gentle learning curve suitable for beginners.

Conclusions:

• The proposed training method effectively introduces data clustering to a broad audience.
• Visual and problem-based learning enhances understanding and engagement.
• This approach lowers the barrier to entry for data science education.