Harris Lin is a data scientist from Plant and Food Research, whose team has expertise in computer vision technologies, signal processing from smart sensors, natural language processing, data-driven consumer insights, and genomic data science. He is passionate about building predictive model pipelines and MLOps for various computer vision and other data science projects. A Software Engineering alumni of University of Auckland, Harris received his Ph.D. from Iowa State University, and spent a couple of years working in the cyber security industry before joining Plant and Food Research.
Machine learning, or predictive modelling in general, has found its way of transforming our lives one way or another in many applications. However in some cases even scientists can be easily misled by a flood of buzzwords and unfounded claims. This talk will give a set of recommended measuring sticks for assessing a predictive model. It will also cover guidelines for deploying and monitoring a predictive model in production, instead of using and trusting its predictions forever. In this talk I will try to demonstrate these concepts with examples of our past and current machine learning and computer vision projects in Plant and Food Research.
If you want to join this seminar, please leave your name and email below, you will get the Zoom link once you click submit.