The content on this page was translated automatically.
Humboldt Fellow Dr. Flora Dilys Salim researches machine learning
Since August of this year, Dr. Flora Dilys Salim from Australia has been working at the University of Kassel as part of the Humboldt Research Fellowship for Experienced Researchers. Her host is Prof. Dr.-Ing. Klaus David, Head of the Department of Communication Technology. Dr. Salim will be conducting research at the University of Kassel for six months.
Salim's research interests focus on machine learning, including context and behavior modeling, data mining and pattern recognition. Her research is particularly oriented towards digital assistants in the context of the "Internet of Things", for example in apps for more precise everyday planning.
Her project at the University of Kassel looks more closely at intelligent assistants and the challenges they face in making predictions from multimodal sources with minimal user input or interaction due to the situational context. Overall, David and Salim's collaboration focuses on several key areas. One of these is aimed at improving data quality, which is necessary as a basis for machine learning. The method of transfer learning is used for this purpose. Transfer learning uses pre-trained models from other contexts.
Professor David and Dr. Salim have already begun researching a deep learning method for transfer learning with sensor data from several cities in order to accurately predict the optimal room temperature in an office building in the USA.
Dr. Flora Dilys Salim (38) is an Associate Professor in the School of Science at the Royal Melbourne Institute of Technology (RMIT) and Deputy Director of the RMIT Centre for Information Discovery and Data Analytics.