Student Summer Scholarships
This year we are able to offer a number of summer placements within the Centre to outstanding students . Each successful student will work alongside a research supervisor from SUCCESS on their own research project. Depending on the project, students will get the chance to experience internal or external collaborative research projects with industry and research partners.
The Scholarships will run from November/December 2011 to February 2012 (10 weeks, start and finish dates to be finalised with supervisors). Students will receive a tax-free stipend of $4000.
Application and Eligibility:
- Applications should be submitted to the Centre Administrator Gillian Foster, by November 18, 2011.
- Applications should include a CV with course transcript, and statement ranking your preferred topics (see the list below).
- To be eligible to apply, students must be Undergraduate or Masters students from Swinburne with an average mark of 75 or more, and who intend to continue study in 2012.
Any enquiries should be sent to the Summer Scholarship Coordinator, Dr. Alan Colman
Project List:
Project Details:
A mini-system for interpreting terms in keyword queries
Currently, keyword search has been utilized widely in Web Search and Database Query Processing. However, the ambiguity of keyword search intents leaves users unsatisfied with the returned results. To clarify the keyword search intents, we propose to design a mini-system that can generate a set of interpretations for a given query term with regards to the specified dataset, which can be used as a basic support for Web Search Engine and Database Query Processing Engine.
Requirements of Applicants:
Essential Requirements:
- Good programming skills in C++ or Java
- Fluent communications in English
Desirable Requirements:
- Positive attitudes about hard-problems
- Knowledge about XML data
- Keening to pursue a higher degree is a plus
Contact: Dr. Jianxin Li, Professor Chengfei Liu
Proactive Maintenance of Metro Trains
Can be extended to honours project.
This is a joint project between Swinburne University of Technology and Metro Trains. Certain kind of computer system services show failure symptoms or degradation in service level over a period of time until they fail eventually. Examples are the reliability of hardware components, performance of software components, connectivity between data communication equipment and data transmission. We conjecture that similar characteristics exist in the hardware and electronic components of a train. If so, a system that monitors and predicts such failures can provide warning when such a failure trend is detected. It can recommend proactive maintenance to reduce service disruptions and improve the efficiency of train services in Melbourne.
Required Skills: We invite students who have an outstanding academic record and would like to work with real-life problems to join us. The successful candidate must be able to develop complex programs, carry out data analysis and understand statistical testing.
Contact: Dr. Antony Tang
Embedded Systems testing
Embedded systems are computer software systems, which are embedded in electronic devices to perform a few specific tasks. Many embedded systems are life- and mission-critical systems, which require rigorous testing. This project aims to investigate effective testing techniques for embedded systems.
Contact: Dr Fei-Ching Kuo
Enterprise Software Testing with Coloured Petri Nets
Coloured Petri Nets (CPNs) is a language for the modelling and validation of systems in which concurrency, communication, and synchronisation play a major role. Examples of such systems are business process and workflow modelling, manufacturing systems, and agent systems. CPN Tools (http://cpntools.org/) is an industrial-strength graphical tool for constructing, simulating, and analysing CPN models.
This project aims at implementing a Java client to interact with a CPN emulation model of the popular communication protocol LDAP, and integrating this client with an enterprise software testing environment to be applied at CA (Computer Associates) Labs Melbourne.
Contact: Dr Jian Yu
A Model-Driven Tool for Adaptive Context-Aware Mobile Services
Context-aware mobile services are becoming popular with the rapid development of mobile devices, World-Wide Web, and modern component-based software development paradigm. However, current process-based approaches to building components and services are too rigid for context-aware mobile applications because of their dynamic behaviour.
This project aims at implementing a tool to 1) specify the context-aware features of a mobile application using business rules; 2) weave the rules to the process using an aspect-oriented engineering principle; 3) automatically translate the weaved model to an executable language.
Contact: Dr Jian Yu
Simulating Smart Appliances for Intelligent Energy Management
Can be extended to honours project.
Project Task: The Smart Grid refers to a series of initiatives towards the modernization of the current electricity generation, distribution and consumption chain. Among the proposed developments, Smart Meters will provide a fine-grained, two-way communication channel between consumers and electricity companies, enabling them to charge different prices for different times of the day and change these prices dynamically so as to fulfil the companies' goals, such as peak and carbon footprint reduction. To complement this technology, manufacturers are developing Smart Appliances, that can take advantage of this dynamic pricing model, and software companies are developing household portals to enable the interconnection and formation of household networks of Smart Appliances.
This project will develop software simulators of the selected Smart Appliances that can be intelligently controlled and coordinated with dynamic pricing mechanisms. The simulated Smart Appliances will be interfaced and demonstrated with the intelligent energy monitoring and control devices and portal at the Energy Management Research Centre at Swinburne.
Good modelling, programming and software engineering skills are required.
Contact: Prof Ryszard Kowalczy, Dr Bao Vo
Interfacing Smart Energy Management at Swinburne
Can be extended to honours project.
Project Task: This project will involve investigation and development of a data acquisition system interfacing with an existing Building Energy Management system, and integrating it with the monitoring and control infrastructure at Energy Management Research Centre at Swinburne.
Contact: Prof Ryszard Kowalczyk, Dr Bao Vo
Negotiation Web Service
Can be extended to honours project.
Project Task: "Negotiation is a process of interaction between parties directed at reaching some form of agreement that will hold and which is based upon common interests, with the purpose of resolving conflict, despite widely dividing differences". When negotiation is carried out by autonomous software entities, it constitutes an autonomous negotiation. The Intelligent Agent Technology (IAT) research group has been working with this technology and developed a series of different mechanisms and strategies over the years. The variety of approaches has lead to a series of disparate implementations that now need to be put together for the benefit of the community.
This project will develop a library of Negotiation Web Services. It will involve re-design and implementation of different negotiation mechanisms and strategies developed by the IAT researchers. The successful candidate will have the opportunity to work with advanced techniques of Artificial Intelligence and a broad range of programming languages and models (Java, Mathematica, etc). Good analytical, programming and software engineering skills are required.
Contact: Prof Ryszard Kowalczyk, Dr Bao Vo
Smart Cloud Broker for Infrastructure as a Service
Can be extended to honours project.
Project Task: Infrastructure as a Service (IaaS) is a cloud service that allows organizations to outsource the computing resources they use, including storage, hardware, servers and networking components, to the cloud. Gartner Research states that "the future of cloud computing will be permeated with the notion of brokers negotiating relationships between providers of cloud services and the service consumers". With the growing number of IaaS providers and provisioning models, a cloud broker can adopt innovative business models on top of this. This can vary from a simple 'IaaS selector' which matches consumers requirements to cloud provider capabilities, to a more complex 'IaaS reseller' where the IaaS Broker can rent resources from various cloud providers under different pricing models, and then resell them to end-consumers with the objective of making a profit.
This project will involve the research and prototypical development of a new cloud brokering service. It will include a study of the different IaaS provisioning models at the existing cloud management platforms, exploration the different business models for the cloud brokering service, and its demonstration by extending the Smart Cloud Broker prototype developed at Swinburne.
Contact: Mohan Baruwal Chhetri, Dr Bao Vo, Prof Ryszard Kowalczyk
Benchmarking the Cloud
Cloud computing is widely considered to be the next generational step in Information and Communication Technology, and it is being intensively investigated and applied by research institutes, enterprises and government agencies. Currently, there are many public cloud service providers such as Amazon, Google, IBM, Saleforce.com, Rackspace, and Microsoft, but the features of their services such as the types, prices, functionalities, performance, scalability and reliability are very different from each other. Therefore, to conduct a comprehensive survey on major cloud services, i.e. "benchmarking the cloud", is a timely issue for the research and development of cloud computing applications.
This project aims to survey and collect the performance data of public cloud services such as Amazon EC2, S3, CloudFront, and competitive counterparts of other major service providers. For such a purpose, a set of testing scenarios, benchmarking data and applications need to be designed and implemented using commonly supported programming languages such as Java and PHP with the provided API tools to program the cloud services. The benchmarking tools and survey data produced by this project will contribute to the future cloud computing projects carried out in the faculty, and the student who undertakes this project will have hands-on experiences with cloud computing and master sufficient skills for programming different types of cloud services.
Contact: Dr. Xiao Liu, Dr. Alan Colman, Prof. Yun Yang
Data reliability management with proactive data loss detection in Cloud Computing
Can be extended to Honours Project
With the development of the Cloud computing paradigm, Cloud-based applications have put forward higher demand for Cloud storage: Data in the Cloud need to be stored with higher cost-effectiveness. By reducing data replicas stored in the Cloud, the storage space consumption and storage cost can be controlled. However, the data reliability requirement is hard to be met at the meantime.
This project aims to implement a prototype of a reliability management system based on Amazon Web Services. This system is able to provide several major functionalities of data reliability management including data upload, data replication, proactive data detection, failure control. By applying a novel Proactive Detection strategy, the purposes of data reliability management and cost-effective data storage can both be met.
Good Java programming skill is required. Familiarity with Eclipse IDE is desirable.
Contact: Prof. Yun Yang, Mr Wenhao Li
Efficient Data Storage and Retrieval in Large-Scale, Distributed Software Systems
Can be extended to Honours Project
Contemporary software systems are increasingly deployed into cloud environments, characterized by their large-scale, heterogeneous and distributed nature. In an effort to further the understanding of issues in deploying software systems into such environments, we are currently developing an emulator capable of masquerading as such an environment for a variety of purposes. One challenge in developing the emulator is to efficiently map persistent data of large-scale, complex environments consisting of tens of
The main goal of this project is to define and implement models suitable for the efficient storage and timely retrieval of node data. More specifically, we are interested in exploring state-of-the-art techniques such as data deduplication, copy-on-write, and lazy data production for our emulator. It is expected that the results of this project will be incorporated into our emulator, furthering our investigations into large-scale distributed environments.
The work will form part of a larger project conducted in collaboration with our industry partner CA technologies and will give interested students insights into practices, tools and techniques used in industry. Furthermore, interested students can use this project to familiarize themselves with the general topic area of cloud environments as a stepping stone towards a potential Honours project in 2012 or beyond.
Required Skills: interested students should have a working knowledge of object-oriented concepts and good programming skills (in at least one object-oriented language). Familiarity with basic mathematical notations and concepts is a plus.
Contact: Dr. Jean-Guy Schneider, Cameron Hine
Spectrum-Based Fault Localization
This project aims to conduct an experimental analysis of applying spectrum-based fault localization in the absence of testing oracle. Applicants should have basic knowledge of software engineering, and good programming language skills.
Contact: Prof T. Y. Chen, Xiaoyuan Xie
Intelligent Folder
Self-Organizing Maps (SOM) are artificial neural networks that transform high dimensional input data into low dimensional data representation such as a 2 dimensional map. This project will investigate, implement and evaluate SOM for classifying text documents. The aim is to explore the feasibility of implementing an "intelligent" folder, where one can drag documents into the folder and it automatically puts the document in an appropriate sub folder.
Candidates should have strong programming background and a keen interest in neural network and natural language processing.
Contact: Dr Caslon Chua
Document Forensic
One of the major tasks in natural language processing is automatic summarisation. This project is to investigate and utilise existing natural language processing tools to extract key words/phrases and/or summarise text documents. The aim is to explore the possibility of clustering documents by topics and/or purpose.
Candidates should have strong programming background and a keen interest in natural language processing and AI.
Contact: Dr Caslon Chua
Exploring the best trade-off between computation and storage in cloud computing
Can be extended to honours project.
Cloud computing has emerged as a promising solution for the computation and data intensive applications. Given the excessive processing power and storage capacity of cloud computing system, large applications data can be either stored in the cloud or re-compute whenever needed. Due to the pay-as-you-go model (usually adopted by commercial cloud service providers), there is a trade-of between computation and storage in the cloud.
This project aims to find the best trade of between computation and storage, which is the theoretical minimum cost for deploying applications in the cloud with the pay-as-you-go model. This minimum cost can be utilised as a benchmark to evaluate the cost-effectiveness of all other data storage strategies in the cloud. In this project a real data-intensive scientific application is investigated as an example, and Amazon cloud services' pricing model will be adapted in the experiment for evaluation.
Java programming skill is required.
Contact: Mr. Dong Yuan, Prof. Yun Yang
