Dr Ke Deng
Dr. Deng was awarded a PhD degree in computer science at The University of Queensland (UQ) in 2007 with focus on data and knowledge engineering. He also hold a Master degree in Information and Communication Technology in 2001 and a Bachelor degree in Electrical Engineering in 1994. He had been an acting lecturer and course convener in the School of Information Technology and Electrical Engineering at UQ during 2008-2012, and the co-supervisor of one PhD student. He had been a postdoctoral research fellow at CSIRO ICT center in 2007. He had been an ARC Australian Postdoctoral Fellow working during 2010-2012. He had been a researcher in Huawei Noah Ark’s Research Lab, Hong Kong from June 2013. The Noah’s Ark Lab conducts state-of-the-art research in data mining and artificial intelligence by exploring theories and building intelligent systems.
More information on Dr Deng's research interests and publications can be found on his personal website.
- COSC1114 Operating System Principles
- 2015 PhD Project (one scholarship, please contact me if you are interested)
The smart city are also known as ubiquitous city, intelligent cities, virtual cities, digital cities, knowledge cities which are all perspectives on the idea that ICT is central to the operation of the future city, even though it has been recognized the use of ICT alone does not make cities smart - other dimensions include cultural and natural amenities, people and skills, knowledge precincts and governance. Many cities in Asian, Europe, and America such as Hong Kong, Amsterdam and Osaka currently have smart city projects. Robust urban computing requires the combination of a huge amount of static knowledge about the city (i.e. urban, social and cultural knowledge) with an even much larger set of dynamic data (originating in real time from heterogeneous and multiple sources). The most dominant dynamic data perhaps are the trajectories of massive humans and vehicles in the city. In the last decade, trajectory-data-intensive urban computing has attracted much attention from ICT researchers and industrial practitioners to address problems, combined with other city information, in urban planning, city transport, social applications, environment, urban energy consumption, economy, public safety and security.
However, urban computing on trajectory data is still in its early stage. First, the concept of smart city encompasses so many aspects of complex city life such that new urban computing problems continuously emerge, for example, the recharge station positioning for electrical vehicles. Second, most exiting urban computing solutions explore single-source data (e.g., using taxi trajectory data only); as a consequence, the robustness of solution is largely constrained by the limited information of the single-source data (e.g., taxi trajectory data has nothing related to the population using public transport systems). In the progress towards smart city, the information about different facets in the same urban space, unknown previously, becomes more and more available, for example, the trajectories of bus/subway passengers using smartcard ticketing system devices, the trajectories of cell phone users, the trajectories of buses, and the trajectories of social media check-ins. While heterogeneous trajectory data provide unprecedented rich information to enable robust solutions of urban computing problems, it introduces enormous technical challenges in order to settle heterogeneity.
Aim 1: Develop a heterogeneous data management system which comprises a distributed data storage and management layer and data access layer.
Aim 2: The comprehensive transport patterns in hierarchical scale levels will be disclosed to enable robust traffic efficiency analysis throughout a city, to learn the impact of city events to local traffic, and to enable other urban computing tasks investigated in this project.
Aim 3: The study provides insight to the characteristics of regions and social communities in the city which helps better facility planning, social interaction, and collaboration of local communities for mutual benefits.
- Masters Project
- Xu, X.,Deng, K.,Zhang, X. (2022). Identifying Cost-effective Debunkers for Multi-stage Fake News Mitigation Campaigns In: Proceedings of the 15th ACM International Conference on Web Search and Data Mining (WSDM 2022), Tempe, Arizona, United States, 21-25 February 2022
- Zhang, M.,Wang, G.,Ren, L.,Li, J.,Deng, K.,Zhang, B. (2022). METoNR: A meta explanation triplet oriented news recommendation model In: Knowledge-Based Systems, 238, 1 - 12
- Wang, J.,Ren, Y.,Li, J.,Deng, K. (2022). The Footprint of Factorization Models and Their Applications in Collaborative Filtering In: ACM Transactions on Information Systems, 40, 1 - 32
- Zhao, H.,Pang, C.,Kotagiri, R.,Pang, C.,Deng, K.,Yang, J.,Li, T. (2022). An Optimal Online Semi-Connected PLA Algorithm with Maximum Error Bound In: IEEE Transactions on Knowledge and Data Engineering, 34, 164 - 177
- Wang, A.,Wen, S.,Gao, Y.,Li, Q.,Deng, K.,Pang, C. (2022). Toward Enhancing Room Layout Estimation by Feature Pyramid Networks In: Data Science and Engineering, 7, 213 - 224
- Zhang, M.,Liu, J.,Zhang, W.,Deng, K.,Dong, H.,Liu, Y. (2021). CSSR: A Context-Aware Sequential Software Service Recommendation Model In: Proceedings of the 19th International Conference on Service-Oriented Computing (ICSOC 2021), Virtual Event, 22-25 November 2021
- Wang, A.,Wen, S.,Gao, Y.,Li, Q.,Deng, K.,Pang, C. (2021). An Efficient Method for Indoor Layout Estimation with FPN In: Proceedings of the 22nd International Conference on Web Information Systems Engineering (WISE 2021), Melbourne, Australia, 26-29 October 2021
- Zhang, M.,Yang, Y.,Abbas, R.,Deng, K.,Li, J.,Zhang, B. (2021). SNPR: A Serendipity-Oriented Next POI Recommendation Model In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management (CIKM '21), Virtual Event, 1/11/2021-5/11/2021
- Balasubramanian, L.,Wurst, J.,Botsch, M.,Deng, K. (2021). Traffic Scenario Clustering by Iterative Optimisation of Self-Supervised Networks Using a Random Forest Activation Pattern Similarity In: 2021 IEEE Intelligent Vehicles Symposium (IV), Nagoya, Japan, 11/07/2021 - 17/07/2021
- Balasubramanian, L.,Kruber, F.,Botsch, M.,Deng, K. (2021). Open-Set Recognition based on the Combination of Deep Learning and Ensemble Method for Detecting Unknown Traffic Scenarios In: 2021 IEEE Intelligent Vehicles Symposium (IV), Nagoya, Japan, 11/07/2021 - 17/07/2021
10 PhD Current Supervisions and 1 Masters by Research Current Supervisions6 PhD Completions
- Stream Data Classification in the Age of 5G Networks. Funded by: ARC Discovery Projects 2021 from (2021 to 2024)
- Personalised Online Learning Analytics by Exploring Multilayer Graph Data (Administered by Deakin University). Funded by: ARC Linkage via Other University from (2019 to 2021)
- Effective and Efficient Query Processing over Dynamic Social Networks (Administered by Swinburne University of Technology). Funded by: ARC Discovery Projects 2016 from (2016 to 2019)