Machine Learning – Case Studies in Biological Sciences

Ms Usha Nattala1

1Melbourne Data Analytics Platform, University Of Melbourne, Melbourne, Australia

Abstract:

INTRODUCTION

Machine learning is a new and exciting field that has just begun getting applied to diverse non-computer science fields such as biosciences, medicine, veterinary and agricultural sciences which traditionally relied on statistical methods to drive research outputs. Several of these efforts are an Australian-first, leading the way for the others in the field and making possible what was previously considered impossible.

Each of these fields has its own unique set of challenges, pain points and preferred ways of tackling limitations. Ranging from limited training data to vague specifications, every machine learning project we tackled was an intense journey of learning and growth.

METHODS

I would like to present some of our machine learning case studies to help future researchers learn from our experiences:

  1. Pollen Monitoring and Prediction (Botany)
  2. Acoustic analysis of speech changes with neuro-degenerative disease (Neurosciences)
  3. Nitrogen cycle tracking for agro-ecosystem modelling (Agricultural Sciences)
  4. Racehorse injury and fatality prediction (Veterinary Sciences)

RESULTS

We will examine the inputs and data cleaning performed in each case, the algorithms employed, the success ratio and validity of the obtained research results using these methods.

CONCLUSION

Machine learning can bring a real value-add to these fields which have just begun benefiting from the recent advances in data and computational sciences.


Biography:

Usha is a research data specialist and software developer with over a decade of experience across multiple technology stacks. Specializing in application development, data science, machine learning and math-intensive programming, she quite enjoys the process of bringing to life the ideas of the best and brightest minds of our generation.

Usha works at the Melbourne Data Analytics Platform (MDAP), University of Melbourne. Conceptualized under the Petascale Campus Initiative (PCI), MDAP is the university’s premium data and computational science team helping researchers across the community in the fields of high performance computing, big data, machine learning, natural language processing and text mining, analysis and visualisation, virtual and augmented reality, image and video processing, scientific simulation and modelling, and web application development.

ABOUT AeRO

AeRO is the industry association focused on eResearch in Australasia. We play a critical coordination role for our members, who are actively transforming research via Information Technology. Organisations join AeRO to advance their own capabilities and services, to collaborate and to network with peers. AeRO believes researchers and the sector significantly benefit from greater communication, coordination and sharing among the increasingly different and evolving service providers.