Data Mining, Visualization & Predictive Analytics using Orange
A hands-on, no-code introduction to the full analytics pipeline — from raw data to a working predictive model — built entirely on Orange, the free open-source visual programming tool trusted by researchers worldwide.
Every project in Orange is a canvas of connected widgets — the same logic we'll build live in the workshop.
Free & open source, always. Developed by the Bioinformatics Lab at the University of Ljubljana and released under an open-source licence, Orange carries no licence fee, no seat limits, and no vendor lock-in — ideal for students, departments, and grant-funded research alike.
What is Orange, and why build your analytics on it?
Orange is a visual programming platform for data mining and machine learning. Instead of writing code, you drag widgets onto a canvas and connect them into a pipeline — each widget doing one job, from loading a file to training a classifier.
-
🧩
Drag-and-drop workflows
Build a full data-mining pipeline — import, clean, visualize, model, evaluate — without writing a single line of code, and see results update live as you connect widgets.
-
🔓
Genuinely open source
The full source is public and community-maintained, so workflows are transparent, auditable, and reproducible — exactly what peer review and replication demand.
-
🧬
Extensible with add-ons
Purpose-built add-ons cover text mining, bioinformatics, network analysis, spectroscopy, and time series — so one tool grows with your research question instead of forcing you to switch platforms.
-
🐍
Built on Python's data-science stack
Under the hood, Orange runs on the same libraries (NumPy, scikit-learn) researchers already trust — and a built-in Python scripting widget lets you drop into code exactly when you need to.
-
🖥️
Cross-platform & classroom-friendly
Runs the same way on Windows, macOS, and Linux, with a visual canvas that's far easier to teach, demonstrate, and grade than a script full of syntax errors.
Why it matters for researchers: Orange lowers the barrier between "I have data" and "I have a defensible model." You can explore distributions, test preprocessing choices, compare classifiers side by side, and share the exact workflow with a co-author or reviewer — all without a programming background, and all for free.
Who gets the most out of it
Research scholars & PhD students
Analyse survey, experimental, or secondary data and build models for your thesis or publication.
Faculty & academicians
Bring a no-code, visual way to teach data mining and machine learning in the classroom.
UG / PG students
Statistics, computer science, and management students building a practical analytics foundation.
Industry analysts
Professionals who want a fast, low-code route into predictive analytics without learning to code first.
Absolute beginners
No prior programming or statistics background needed — Orange's visual canvas is the on-ramp.
Interdisciplinary researchers
Anyone whose field increasingly demands data literacy but not necessarily a coding background.
What you'll learn, day by day
- 01Data mining fundamentals
- 02The data mining life cycle
- 03Data collection methods
- 04Sampling techniques
- 05Data cleaning & preprocessing
- 01Data visualization in Orange
- 02Building predictive models
- 03Evaluating model performance
Resource person
Dr. Shanti Verma
Dr. Verma brings hands-on academic and research experience in data mining and analytics to a practical, workshop-style format — built around live demonstrations in Orange rather than slides alone.
Certificate & recordings
E-certificate
Issued to every registered participant after the workshop concludes.
Session recordings
Both days are recorded and shared afterwards, so you can revisit any concept.
Live, hands-on format
Real-time demonstrations in Orange over Zoom, not a pre-recorded course.
Reserve your spot for 7–8 September 2026
Choose the registration option that matches where you're joining from.