Overview
Learn how to mine data using Weka, with the University of Waikato
On this five-week course, you’ll discover how to mine data using the Weka workbench, a powerful tool for machine learning and data mining.
Guided by experts at the University of Waikato, the original developers of Weko, you’ll learn the basics of data visualisation, classification algorithms, and data interpretation and evaluation.
Explore the basics of data interpretation and evaluation
Beginning with an introduction to data mining concepts, you’ll discover the various applications of data mining in personal and professional contexts.
You’ll examine how to evaluate a classifier’s performance and use training, testing, and cross-validation to gauge the accuracy of the data you’ve gathered.
With these skills, you’ll be able to improve the quality of your data and develop meaningful answers to the questions you’re trying to answer.
Organise your data using classifiers
Exploring both simple and more complex classifiers, you’ll learn how different classification methods can be used to interpret datasets.
You’ll investigate the applications of concepts including decision trees, linear regression, and support vector machines, learning how to apply the correct classification method to your problem.
Examine the full data mining process
In the final week of this course, you’ll put your learning into context by exploring the full data mining process.
You’ll address common pitfalls and challenges to accessing data, as well as assessing the ethics of data mining, giving you a broader understanding of how and when data mining should be used in different contexts.
You’ll finish this course understanding what Weka is and how to gather and interpret big data. You’ll be aware of the full data mining process and be able to explain and apply Weka within your own data mining work.
This course is designed for anyone considering a career in data science or those currently working in the data sector wanting to further their knowledge of data mining software.
You will download the free Weka software during Week 1. It runs on any computer, under Windows, Linux, or Mac. It has been downloaded millions of times and is being used all around the world.
(Note: Depending on your computer and system version, you may need admin access to install Weka.)
Syllabus
- A little bit of everything
- What’s data mining? What’s Weka? What’s the course about?
- What’s it like to do data mining?
- Exploring the Explorer
- Exploring datasets
- Building a classifier
- Using a filter
- Visualizing your data
- Evaluation
- How do I evaluate a classifier’s performance?
- Be a classifier!
- Training and testing
- Repeated training and testing
- Baseline accuracy
- Cross-validation
- Cross-validation results
- How are you getting on?
- Simple classifiers
- How do simple classification methods work?
- Simplicity first
- Overfitting
- Using probabilities
- Decision trees
- Pruning decision trees
- Nearest neighbor
- More classifiers
- What about real-life classification methods?
- Classification boundaries
- Linear regression
- Classification by regression
- Logistic regression
- Support vector machines
- Ensemble learning
- Putting it all together
- What else is there to know?
- The data mining process
- Pitfalls and pratfalls
- Data mining and ethics
- There’s no magic in data mining
- Farewell