There are several applications of machine learning in the real world. Some of which are medical diagnoses and even detection of contaminants in soil samples.
Add to that the project of researchers that involves investigating whether machine learning can be used to protect New Zealand’s biosecurity.
Pest species are a particular threat to the country’s unique biodiversity. However, the question in mind is how to differentiate whether a species is a pest or merely a harmless member of the local fauna and flora.
According to a recent press release, researchers from the University of Waikato and Canterbury University are working on a program to help everyday Kiwis identify pest species with the use of their smartphone.
How machine learning can help?
A Professor of Computer Science from the University of Waikato will discuss some of the underlying machine learning technology during a public lecture, which will draw connections to work on statistical species identification that pre-dates the computer age.
Additionally, he will address how machine learning can be used to identify species automatically, by learning from photos that have been labelled by experts.
The research focus of Professor Elbe Frank includes machine learning, data mining and artificial intelligence (AI) and their application for the real world.
He was instrumental in the development of WEKA, the University’s award-winning open-source machine learning software platform.
According to him, learning to discriminate species from data is one of the oldest, if not the oldest, application of machine learning.
Ronald Fisher, a statistician and biologist, described a method for linear discriminant analysis in 1936 and applied it to the classification of species of iris flowers.
His method is now a classic technique for supervised machine learning – learning from expert-labelled observations – but it was published years before computer scientist Alan Turing discussed the idea of learning machines in his seminal 1950 paper on ‘Computing Machinery and Intelligence’.
Artificial neural networks
He explained that recent developments in the field of artificial neural networks, which is sometimes referred to as ‘deep learning’, will open up new opportunities for automatic species identification based on photos taken with digital cameras.
Artificial neural networks can automatically learn a set of features to represent images, often yielding more accurate image classifiers than those based on ‘hand-crafted’ image features.
AI can often be overhyped or oversold, but there are many applications where machine learning can improve outcomes for people and increase productivity for organisations.
Although New Zealand is lacking university graduates with significant knowledge in this area, the Professor and his colleagues in the Waikato Machine Learning Group are working hard to change this.
The lecture, called ‘Learning to discriminate species from data: Then and now’, is part of the University’s Hamilton Public Lecture Series and is free and open to the public.