Flora Salim

Flora Salim is a Professor in the School of Computing Technologies, the co-Deputy Director of RMIT Centre for Information Discovery and Data Analytics (CIDDA), and an Associate Investigator of ARC Centre of Excellence in Automated Decision Making and Society. Flora leads the Context Recognition and Urban Intelligence (CRUISE) group.

Flora's research interests include machine learning on stream and sensor data, behaviour modelling, time-series and spatio-temporal data mining, mobility data science, and ubiquitous computing. She has received more than $10M in research funding in the last 10 years. Her research has been funded by Australian Research Council (ARC), Victorian Government, Microsoft Research, Northrop Grumman Corporation US, Rheinmetall Defence Australia, Qatar National Priorities Research Program, IBM Research, Alexander von Humboldt Foundation, Bayer Foundation, city councils, and several other industry and local government partners.

She was a Humboldt-Bayer Fellow, Humboldt Fellow (experienced researcher), Victoria Fellow, the recipient of the RMIT Vice-Chancellor's Award for Research Excellence–Early Career Researcher 2016; the RMIT Award for Research Impact - Technology 2018; the RMIT School of Science HDR Supervision Excellence Award 2017; Australian Research Council (ARC) Postdoctoral Research Industry (APDI) Fellow (2012-2015), and IBM Smarter Planet Industry Skills Innovation Award (2010).

She serves as Associate Editor of the PACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT), Area Editor of Pervasive and Mobile Computing, and a Steering Committee member of ACM UbiComp. She was a Visiting Professor at University of Kassel, Germany, and University of Cambridge, England, in 2019.

Industry experience

Senior software engineer, mediaproxy and Xenon Systems

Research

Her expertise lies at the intersection of ubiquitous computing (on capturing and modelling human mobility and dynamic user behaviours) and data science and machine learning (particularly on time-series, spatiotemporal, and multimodal sensor data). She has many years of experience in analysing human behaviours in multiple contexts and developing key techniques to enable situational awareness for stakeholders and personalised intelligent assistance for the end users. She always take user-centric approaches of data capture and analysis in modelling and profiling human behaviours. Given the nexus of the two research areas, her research contributions on human behaviour (including mobility) modelling have been largely published in the pervasive/ubiquitous computing venues (including UbiComp/IMWUT, PerCom, Pervasive and Mobile Computing, and IEEE IoT Journal), and the contributions in machine learning with time-series, spatio-temporal, and trajectory data have been published in NeurIPS, WWW, KDD, WSDM, TKDE, PAKDD, and many more.

Research keywords

Machine learning, Deep learning, Time-series, Sensor data, Spatio-temporal data, Ubiquitous computing, Mobility, Data science, Behaviour modelling, Human activity Recognition, Trajectory, Forecasting, Predictive analytics, Emotion sensing, Wearable computing, Trustworthy AI, Fairness in machine learning, Explainable AI

Research output summary

150+

Publications

$10 million+

Projects

Awards

Supervisor interest areas

Deep learning for time-series and sensor data
Large-scale/urban-scale event detection and forecasting with heterogeneous time-series for mobility and related behaviours (e.g. energy, public health)
Fair and explainable machine learning
Human behaviour, emotion, health, wellbeing sensing with wearables
Personalisation and mobile-based and POI recommender systems

Supervisor projects

Self-supervised learning for multimodal data
Forecasting with heterogeneous time-series
Representation learning with uncertainty
Disentanglement representation learning for spatio-temporal data
Spatio-temporal representation learning for situational awareness and traffic flow forecasting
Behaviour and emotion sensing with wearables
Graph and sequential embedding learning
Mobile information behaviour and intent modelling

Feature publications

MobTCast: Leveraging auxiliary trajectory forecasting for human mobility prediction

Advances in Neural Information Processing Systems (NeurIPS)

H. Xue, F. D. Salim, Y.Ren, and N.Oliver. (2021).

Exploring self-supervised representation ensembles for covid-19 cough classification

ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) 2021, pp. 1944-1952

H. Xue and F. D. Salim. (2021).

Time-series change point detection with self-supervised contrastive predictive coding

Proceedings of the Web Conference (WWW 2021)

S. Deldari, D. V. Smith, H. Xue, and F. D. Salim. (2021).