Our Research

We are continually researching and implementing world-first ideas. Read below to see our latest research articles that run the PLATO engine!

PLATO is at the cutting edge of educational innovation. We use technology to empower our students and teachers to be the best they possibly can!

A new look at Cryptocurrencies

Have you ever heard of Bitcoin? It's changing the world as we speak! The randomness of Bitcoin is extremely hard to measure, but we've managed to mathematically define the functions which govern Bitcoin randomness! It turns out that randomness can be measured across a lot of applications, such as education! By using statistical models, we are able to measure a student's progress over time, and provide a completely personalised solution.

On long memory effects in the volatility measure of cryptocurrencies

Did you know that randomness actually can be measured? That's right! Over a while, randomness isn't so random any longer and can be measured. For example, looking at a student's homework set may contain seemingly random errors. However, when you take a look at the trend in that students mistakes over time, a lot of obvious patterns start to appear! PLATO measures, records, understands and can therefore provide personalised solutions to personalised problems!

On generalized bivariate Student-t Gegenbauer long memory stochastic volatility models with leverage: Bayesian forecasting of cryptocurrencies with a focus on Bitcoin

Being able to measure such trends over time, and the patterns that start to become obvious in student learning paths turns out to be a huge integration problem! In high school, we learn about integration in 1 dimension (usually with respect to x). But can you imagine integrating with up to 2,000 dimensions? At once?

Bayesian estimation of Gegenbauer long memory processes with stochastic volatility: methods and applications

Accuracy is king! In order to make sure our findings were valid for many different scenarios, we tested it vigorously in a wide variety of situations. We even generated fake data that we knew the answer to, in order to see how our models behave! The accuracy? Greater than 99%!

On Gegenbauer long memory stochastic volatility models: A Bayesian Markov chain Monte Carlo approach with applications

Our founder Dr. AP submitted his thesis to the University of Sydney in exactly 3 years with 4 published articles. His thesis summarises all the main results, and is a cornerstone for the PLATO engine! By using statistical methods, we can propel students to be the best they possibly can be!