Details of projects currently being undertaken by the RMIT Data Analytics Lab.
Efficient and effective ad-hoc search using structured and unstructured geospatial information
Funding Source: ARC
Staff and students: Timos Sellis, Shane Culpepper, Farhana Murtaza Choudhury, Xiaolu Lu
Web search is a key enabling technology in the information age. However two technologies, ubiquitous mobile devices and massive structured data repositories such as those used to maintain social networking sites, are changing user expectations about how and what should be searched. A key challenge in the research community is how to integrate structured and unstructured information to improve the quality of search. In this ARC Discovery project, we propose new approaches to ranked retrieval for location-aware search. In particular, we present a plan to combine state-of-the-art research from two domains: spatial keyword search in databases, and ad-hoc search in Information Retrieval to improve the quality of search results.
Searching, combining and interpreting tabular data in scientific papers
Funding Source: Google (Google Faculty Research Awards)
Staff and students: Timos Sellis, James Thom
Scientific papers often contain lots of interesting data in a variety of tabular formats. Finding and being able to combine tabular data from different sources has many benefits as Google Fusion Tables show for general data extracted from sources on the World Wide Web.
Knowledge of the semantics of the tables (and columns and rows within tables) is essential to assist users in being able to accurately and sensibly combine data from different sources.
This project will develop and evaluate techniques that assist users finding, combining and interpreting data from tables in scientific papers.
TRacking Indoor Information BEhaviour (TRIIBE)
Funding Source: ARC
Staff and students: Mark Sanderson, Flora Salim, Yongli Ren (postdoc), PhD students
Partners (external): Viocorp Systems Intelligence Pty Ltd
This ARC Linkage project will research the passive tracking of user’s mobile devices in indoor spaces correlating their spatial behaviour with their information needs to deliver personalised information. The project will create a system that enables owners of large buildings (for example, shopping malls, airports, universities) to better manage their spaces and services.
Biomedical Informatics for Effective Health Decision Making
Duration: 11 months (Feb-Dec 2015)
Funding Source: Victorian Dept of Health (via subcontract agreement with University of Melbourne)
Staff and students: Lawrence Cavedon, Simon Kocbek (postdoc)
Partners (external): Victorian Dept of Health; The University of Melbourne. Other partners (i.e., a hospital) to be decided
Budget: $200,000 overall; $95,000 to RMIT University
The Victorian Department of Health seeks biomedical informatics expertise and skills to retrieve, mine and analyse these datasets. This will be provided through researchers from the University of Melbourne and RMIT University. The researchers will work with the Department’s in-house team and will be expected to be on-site at the Department to access the data two days per week. Research tasks will be identified and prioritised in collaboration with the Department. Learnings from these studies will be used to inform the Victorian Health Innovation and Reform Council (HIRC) on developing a robust data-driven approach to health system planning and management
TransNET for “Integrated Design Infrastructure for Australia's Cities”
Duration: 8 months (Aug 2014 – March 2015)
Funding Source: Australian Urban Research Infrastructure Network (AURIN). This project is part of AURIN Lens 10.
Staff and students: Flora Salim, research assistants
Partners (external): University of Adelaide, University of Melbourne, QUT, UNSW, City of Logan, Renewal SA (South Australia government), City of Melbourne
Budget: $85,000 (RMIT only, total funding secured by the AURIN Lens 10 group: $560,000)
Big data from cities, if managed and visualised well, can be used by authorities, urban designers and planners to inform urban renewal and development projects. The volume and variety of big data present challenges in presenting analysis of collected data for human to cognitively perceive valuable information, particularly in the dynamic and complex urban contexts. This project investigates techniques for aggregating, filtering, analysing, and visualizing multivariate federated data sources. A web-based tool for visualizing different variables in up to four dimensions is already developed and currently being evaluated by urban designers and planners from Adelaide, Brisbane, and Melbourne.
iCO2mmunity: Personal and community monitoring for university-wide engagement towards healthier, more productive, and greener living
Duration: 3 years (July 2014 – July 2017)
Funding Source: RMIT, Victorian Greener Government building, Siemens
Staff and students: Flora Salim, Margaret Hamilton, Jane Burry, 4 PhD students (Amin Sadri, Wei Shao, Irvan Bastian, Hui Song)
This research aims to recognize daily activities and routines and mobility of students and staff in RMIT University based on sensor data from smartphones and mobility data sensed with Bluetooth and WiFi infrastructure. This data will be used in carbon footprint estimation, student activity and performance evaluation, and their health state monitoring.
An Integrated and Real-time Passenger Travel and Public Transport Service Information System
Duration: 3 years (July 2013 – July 2016)
Funding Source: ARC
Staff and students: Athman Bouguettaya, Margaret Hamilton, Flora Salim, Xiaodong Li, Xinghuo Yu, Mohammad Haqqani, Saiedur Rahaman
Partners (external): ARUP
The primary objective of this ARC Linkage project is to design an extensible service oriented architecture to manage legacy databases, applications, and analytical tools, providing a unique framework to link buses, trains, and trams into one uniform and homogeneous transport system. The project will help to provide improved services to the public through a better understanding of journey planning demands in comparison to public transport services. By integrating research through design methods with technological solutions, the project will deliver better quality of service and higher customer satisfaction.
Innovative Road Safety Technologies
Duration: 2 years (July 2013 – July 2015)
Funding Source: Mornington Peninsula Shire (MPS)
Staff and students: Athman Bouguettaya, Timos Sellis, Andy Song, Flora Salim, Manpreet Kaur
Partners (external): Mornington Peninsula Shire
Budget: $75,000 (from MPS)
This project aims to deliver new methods for analysing real-time crowd-sensed data from smartphones to infer road risks from individual and collective driving behaviours.
Effective Question Answer Techniques for Complex Customer Queries
Duration: 12/2014 – 2/2015
Funding Source: Contract research MACE Engineering Group
Staff and students: Jenny Zhang, Mark Sanderson, Lawrence Cavedon, Felix Liu, Yuhang Sun
Partners (external): MACE Engineering Group
We are researching novel question answer approaches to answer short customer queries. We are addressing challenges including searching for questions/answer pairs in several internal and external databases. We are working on a journal paper summarising our findings.
Mining Spatiotemporal Association Rules from Large Scale Transport Schedules
Duration: 11/2014 – 3/2015
Funding Sources: College of SEH seed grant and Americold Group
Staff and students: Andy Song, Jenny Zhang, Junliang Jiang, Chen Liu
Partners (external): Americold Group
We are developing approaches to collect data for GPS transport trajectory logs and investigating approaches to mining spatiotemporal association patterns from the log data. A publication is underway.
Data Mining Complex Transactional and Criminal Networks
Duration: 03/2013 – 03/2016
Funding Source: ARC and Austrac
Staff and students: Xinghuo Yu, Jenny Zhang, Qingmai Wang, David Savage
Partners (external): Austrac
Money laundering, if undetected, poses a major concern for governments and communities. The software system platform for detecting money laundering networks from this project will be the first that can assist intelligence data analysts to detect unknown money laundering networks faster and more accurately, helping fight crimes more efficiently.
User-Adaptive Search and Evaluation for Complex Information-Seeking Tasks
Duration: 2016 – 2019
Funding Source: ARC
Staff and students: Lawrence Cavedon, Mark Sanderson, Falk Scholer, Damiano Spina (postdoc), PhD students
Partners (external): University of Melbourne and SEEK
This project will develop a new evaluation framework to understand and characterize users and their situation within complex, multi-faceted search tasks, exemplified through job-search. User-specific characteristics and situations will be mined from complex profiles and interaction logs for online information services run by the industry partner, SEEK. The new techniques will redefine understanding of task-oriented search, and has the potential to reinvent the user experience for such complex search tasks.