Humboldt Fellow Dr. Flora Dilys Salim Researches Machine Learning

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", which can be found, for example, in apps for more precise everyday planning.

Her project at the University of Kassel deals more in depth with intelligent assistants and their challenges to make predictions through situational context from multimodal sources with minimal user input or interaction. Overall, David and Salim's collaboration focuses on several key areas. One of them aims to improve data quality, which is necessary as a basis for machine learning. For this purpose, the method of so-called transfer learning is applied. Transfer learning uses already pre-trained models from other contexts.

Prof. David and Dr. Salim have already begun research on a deep-learning method for transfer learning using sensor data from multiple cities to accurately predict the optimal room temperature in an office building in the United States.

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 RMIT's Centre for Information Discovery and Data Analytics.

Contact:

Prof. Dr. Klaus David
University of Kassel
FB16, FG Communication Technology
Tel.: 0561 804 - 6341
E-mail: david[at]uni-kassel[dot]de