Machine Learning Group

Machine learning is all about automatically finding useful information hidden in data. The goal is to spot patterns that help us understand what's happening in the real world and make predictions about what might happen next.

Why spend months crafting email filters that spammers will outsmart tomorrow? Machine learning flips the script: feed it examples of spam and legitimate emails, and it discovers the hidden patterns that distinguish them—patterns too subtle or numerous for humans to encode by hand. The same principle transforms retail analytics, where algorithms sift through millions of shopping cart combinations to reveal surprising customer habits that no store manager could spot manually.

With so much raw data available digitally these days, there are many of ways to use machine learning techniques.

Our team works on new, more efficient, algorithms for machine learning and data mining as well as new applications.

We are well known for a software "workbench" called the Waikato Environment for Knowledge Analysis (WEKA), which contains a large number of machine learning techniques. WEKA is widely used for research, teaching, and commercial applications of machine learning.

WEKA applications help dairy farmers decide which cows to keep in their herds, tackling bioinformatics problems like gene interaction discovery, mining supermarket transaction data for high profit product associations, predicting soil chemical levels to guide fertilizer decisions, and processing natural language to extract document keywords.

Further details on Machine Learning Group's website.