- Room No: G.2.11
- Telephone: +64 7 838 4021
- Extension: 8766
- Facsimile: +64 7 838 4155
PhD Topic: "Metalearning and the full model selection problem."
The application of sophisticated machine learning algorithms to enterprise data analysis and to industrial production is becoming increasingly important. As a consequence, people now have to face the new problem of how to choose the best algorithm, or combination of algorithms, to use for a given dataset. This problem is usually termed the full model selection (FMS) problem, which consists of many sub problems, and is an extremely large search space that could be a very time consuming task for humans to explore manually. The goal for this thesis is to design and implement computer algorithms and tools for FMS that not only save human's time, but that also do an equivalent or even better job than a professional data analyst would given the same resource limitations.