|
Vacation Research Experience Scholarships |
||||||||||||||||||||||||||||||
|
This scholarship is for high performing QUT undergraduate students who are interested in having a funded research experience over summer semester. Successful applicants will be working with experienced researchers and will receive a living stipend. To apply you need to:
General InformationThe Queensland University of Technology offers a number of Vacation Research Experience Scholarships in most faculties and centres during the summer vacation period. Offers are made subject to the availability of funding and appropriate supervision. Vacation Research Experience Scholars are normally expected to participate in a project related to a researcher’s ongoing research program. PurposeVacation Research Experience Scholarships provide students with an opportunity to experience research work. This is particularly useful to students considering undertaking postgraduate research towards a higher degree. The scholarships provide contact with active researchers and offer a challenging environment in which students have an opportunity to test for themselves and to demonstrate to others, their potential for research work. DurationThe duration of a Vacation Research Experience Scholarship will be determined by the Faculty and will be specified in the letter of offer. Projects are between six weeks and ten weeks in duration and usually require between 12-15 hours per week. The period of tenure is normally broken into two parts to allow for the Christmas closure of the University and the weekly allowance is not paid during this time. Vacation Research Experience Scholarships may commence between early November and early December and normally conclude by mid March. BenefitsThe scholarship provides a weekly allowance of $155 (tax exempt). ApplicationsStudents who wish to be considered for a Vacation Research Experience Scholarship in 2008/2009 should submit their online application by Friday 3 October 2008. Applicants must ensure their referees submit an email of support for their application into the program to infotech.research@qut.edu.au which clearly states that the reference is in connection with an application for a Vacation Research Experience Scholarship. Applicants for Higher Degree Research Scholarships who also intend to apply for admission to a PhD or Masters by Research program at QUT should note the closing date for applications for the Annual scholarship round is Friday 10 October 2008. Information about how to apply for admission to a PhD or Masters program can be found at http://www.rsc.qut.edu.au/future/apply/ International students applying for a Vacation Research Experience Scholarship must have a current student visa and work permit for the period of the research. For further information, please contact infotech.research@qut.edu.au Available Vacation Research Experience Projects
Investigation of Power-aware computing techniques in sensor networks Survey of Environmental Observatories Investigation of Data Management Techniques in Sensor Networks Modelling and Execution of Dynamic and Exception Workflow for Film Production Processes Generalising an Information Security Analysis Tool Multimedia Information Retrieval: Interactive Mobile Video (Multimedia) Delivery Multimedia Information Retrieval: User-Centred Multimedia Web Search Cyclist/car interaction monitoring Developing the economic potential of Second Life: Business Systems in Virtual Environments Information Retrieval – Query expansion Development of Text Mining System for Mining for Customer Opinions Semantic Annotations to Discovered Knowledge
Investigation of Power-aware computing techniques in sensor networks Abstract: Power and power management is a key issue in the application of sensor networks.We still do not fully understand the relationship between power consumption and sensor performance e.g. recording length, transmission rate etc. This limits our ability to deploy sensor networks that can work over a long period of time.This project will research the current challenges and latest technologies in building power-aware and durable sensor networks. This project can be done with one (1) or more (2-5) students.Depending on the number of students, topics will cover:
Expected Outcomes 1. Each student should deliver a 15-20 page report on literature review and comparative analysis of the leading-edge technologies in the selected field. Supervisors: Dr Jinglan Zhang, Dr Paul Roe
Survey of Environmental Observatories Abstract: Environmental observatories are complex facilities and face a number of challenges in infrastructure and operations. Much research has been devoted in this area around the world.This project will review the leading projects, challenging issues, and state-of-the-art technologies (e.g. service-oriented architecture) in this field. This project can be done with one or more students.Depending on the number of students, topics will cover:
Students with GPA over 6 and good writing skills are preferred. Expected Outcomes: 1. Each student should deliver a 15-20 page report on literature review and comparative analysis of the leading projects and technologies in the selected field.The projects selected should cover those from Australia and overseas countries. Supervisors: Dr Jinglan Zhang, Dr Paul Roe
Investigation of Data Management Techniques in Sensor Networks Abstract: A sensor network for environmental monitoring has been established and more research on data processing will be conducted.A client-server architecture is employed where a PC is used as the server and sensors (3G phones) as clients.Clients collect data, partially process the data and send them to the server over the network for storage and further processing.A simple web-interface has been developed for data access.A systematic investigation on the ways for effective data management in sensor networks should be conducted.A suitable approach chosen for the specific application should be implemented. This project can be done with one (1) or more (2-5) students.Depending on the number of students, topics will cover challenges, techniques, and tools for:
Students with GPA over 6 and good database knowledge and skills are preferred. Expected Outcomes: 1. A 15-20 page literature review report 2. One program which implements one of the most promising way for effective data access. Supervisors: Dr Jinglan Zhang, Dr Paul Roe
Modelling and Execution of Dynamic and Exception Workflow for Film Production Processes Abstract: The screen business comprises all creative and management aspects related to film, television, and new media contents, from concept to production and distribution. It is characterised by business processes with high demands for creativity and flexibility. These processes span a value chain consisting of four major phases: development, pre-production, production and post-production. Whether it is a live action drama with actors, a documentary or animation, the production phase is generally the most expensive in screen business. It includes daily shooting activities over a period varying from a day to several years. At present, shooting is a highly manual activity. It involves processing rather large amounts of data on a daily basis and coordinating many geographically distributed stakeholders, which is time-consuming and error-prone. With such context, the Business Process Management (BPM) Group at Queensland University of Technology, in close collaboration with the Australia Film Radio and Television School (AFTRS), applies BPM principles and technology to optimize and automate film production processes. Despite its potential benefits, the use of BPM for film production is a direction yet to be explored. One of the major challenges hindering the application of BPM systems to this domain is the need for high degree of flexibility. For example, shooting activities, particularly shooting on locations, can be affected by many factors such as weather, location, cast and crew availability, etc. It is quite usual that exceptions and changes occur during production and need to be solved in a timely manner to avoid delays in the schedule. In this project, you are expected to gather and analyse the exceptions and changes that usually occur during a film shooting, and model and execute the exception handling process/workflow using a state-of-the-art BPM system. The outcome of your work will directly contribute to film production process innovation. Pre-requisite Knowledge: Business Process Modelling Supervisor: Dr Chun Ouyang
Generalising an Information Security Analysis Tool Abstract: The Secure Information Flow Analyser (SIFA) is an open-source tool for exploring information flow through networks of communicating components (http://sifa.sourceforge.net/). It works by searching for paths from "high-security" data sources to "low-security" data sinks, for each of the network's typical operating "modes".Modes are currently assumed to be mutually-exclusive, but we recently realised that there would be advantages to allowing some modes to subsume others, so that information flow "between" network modes can be identified too.This project will involve making some extensions to SIFA so that it can be used in this way. Pre-requisite Knowledge: Java programming experience Co-supervisors: Prof Colin Fidge and Dr Diane Corney
Multimedia Information Retrieval: Personalizing Next-Gen News Services using Aggregated Multimedia Contents Abstract: Users increasingly consume news articles on the Web and mobile devices which are enriched by audio, image, and video contents. As broadband Internet is becoming affordable for home users, more-and-more multimedia contents can be delivered to attract users' attention to consume news service. The current approach to generate such contents is still manual, with no technology to automate the multi-modal contents indexing and summarization. To sustain the rapid growth of news contents and overcome limitations of manual work, this project aims to propose the framework, tools and techniques to automatically generate a news service which aggregates multimedia contents for increasing users’ enjoyment. For example, users can browse on the text-with-image topics in order to watch some news video highlights; or alternatively, they can start by watching streaming news video on the Web while the interface provides opportunities to read some overlaid text contents describing further information. Using this approach, the news service becomes more personalized as users have the control over the type of media contents they want to consume, depending on their context of use. Supervisor: Dr. Dian Tjondronegoro
Multimedia Information Retrieval: Interactive Mobile Video (Multimedia) Delivery Abstract: Mobile video interaction increases the necessity for more effective content-based retrieval. Mobile devices have limited capabilities in supporting users to watch the full contents of video due to their small screen size and restricted battery life. Users need to selectively watch particular segments they want to watch to reduce the time and costs of downloading full-video contents. Main tasks: 1) Studying users behaviors, tasks, and activity while interacting with videos on mobile devices; 2) Extraction, indexing, and retrieval of video contents; 3) Smart streaming strategies Supervisor: Dr. Dian Tjondronegoro
Multimedia Information Retrieval: User-Centred Multimedia Web Search Abstract: In 2004 66% of internet users in Australia were already subscribed to high-speed broadband. Information providers are increasingly storing more image, sound, and video content on the Web. However, most of the current Web search engines still mainly support text-based retrieval which is not always useful for day-to-day requirements. Main tasks: 1) Study the behaviour, tasks, and activity of multimedia web search in Australia, 2) Design and develop a prototype that meets the trends and emerging requirements for next-generation Multimedia web search Supervisor: Dr. Dian Tjondronegoro
Abstract: The aim of this project is to develop a library of moduls to enable a hexapod robot to explore its environment with some elementary localization and mapping. At the moment, the robot is only equipped with proprioceptive sensors (that tell the robot the position of the joints of its legs). Basic touch sensors will be added. The project will be conducted in C. Click here for a video of the robot used in this project (Lynxmotion AH3) Supervisor: Dr Frederic Maire
Abstract: The aim of this project is to create a Pyro interface for a hexapod robot using Pyro. Pyro stands for Python Robotics. Pyro provides a programming environment for easily exploring advanced topics in artificial intelligence and robotics without having to worry about the low-level details of the underlying hardware. Click here for a video of the robot used in this project (Lynxmotion AH3) Supervisor: Dr Frederic Maire
Cyclist/car interaction monitoring Abstract: This project aims to automatically analyse the interaction between cyclists and cars on a city road. More precisely, we wish to monitor the relative position of cars with respect to the cyclist in the same lane. This is a computer vision project. You are not assumed to be knowledgeable in image processing techniques, but you should be willing to learn Matlab or OpenCV. Supervisor: Dr Frederic Maire
Developing the economic potential of Second Life: Business Systems in Virtual Environments Abstract: Virtual worlds are now developing into sophisticated environments that are being used by businesses to explore marketing, retail and training applications. As part of the research work in the Business Process Management group, we have developed a piece of custom service software which links a business workflow tool to Second Life, thus enabling the visualisation of business activities within a virtual environment. This means that when business analysts develop models of businesses, they can be viewed within a virtual environment for clients and other stakeholders to verify that the model is correct. It also means that businesses can train their staff in virtual environments, rather than having to use videos, or other costly methods. Many large companies, such as IBM, Coca Cola, Givenchy, Toyota, to name a few, are seeking to understand how best to exploit these new environments for business purposes. A video of the work is shown here on the www.bpmve.org website. The work completed so far (shown in the video) animates business processes running on a workflow server. In this project, you will aim to extend the animation capabilities to include interaction approaches. For instance, you will create and script a 3D environment that embeds the interaction with business software into the objects in the scene. Thus, while a staff member is using objects in SL, they are actually driving a piece of business workflow software. Student Value Proposition: This leading edge research area will position the vacation student with skills that will be of interest to companies looking to exploit these technologies. Gartner, a large IT consulting company, believes that virtual environment technology will emerge as a leading driver of new business IT innovation. This project will position you to understand this technology from a novel business perspective. Skills Required: Programming skills – Java, JavaScript and ability to learn the Second Life Scripting Language; Understanding of Virtual Environments/Games/Second Life – no need to have programmed a game, but need to understand the principles of 3D environments; Understanding of Business Process Management will be useful, but not essential. Supervisor: Dr Ross Brown
Information Retrieval – Query expansion Abstract: Search engines perform searches using keywords entered by a user. Often it is possible to obtain better results in specific domains by using query expansion. Examples of query expansion are the use of synonyms or lookup of tables of related terms. In specialised domains, for instance in medicine, it is possible to add terms to a query with confidence. For instance, a query on “mad cow disease” may be expanded to include “bovine spongiform encephalopathy” or “bse”. The objective of the project is to develop a program that given a set of terms will access various web sites to conduct a preliminary search. Using the top results from numerous sources the program will then collate the information, classify it into several possible contexts, and produce a set of suggested expansions for the original query. Implementation environnent: Java, JavaScript, Firefox, Windows/Linux Good programming skills and enjoyment of programming are essential. Supervisor: Dr Shlomo Geva
Abstract: The project involves development of an enhanced Wikipedia Browser. The standard browser is rather limited and in particular does not support the navigation of an extensively linked web of related topics. A user who explores the Wikipedia has to navigate manually through links, move forward and backwards using only the standard and rather limited navigation features of a browser (e.g. use forward/backward or history navigation, or use multiple tabs, etc.) We are looking to create a Wikipedia browser that will be dedicated to the interaction mode of wikipedia users. It should provide a totally different exploration experience. It should support multiple links per anchor, it should remember where a user had been and provide a suitable visual representation to assist the user. It should try and predict what page a user might visit next, or suggest other links. It should assist users in discovering related documents that are not explicitly linked to the viewed document, and so on Implementation environment: Java, JavaScript, Firefox, Windows/Linux Good programming skills and enjoyment of programming are essential. Supervisor: Dr Shlomo Geva
Abstract: The emergence of multicore CPUs has thrust parallel computing into the mainstream. This project will investigate how best to exploit multi-core CPUs to improve the performance of general purpose desktop and server applications, while keeping the development process relatively simple. Specifically, the project will examine the overheads associated with running multithreaded parallel codes via different languages and runtime systems. This will include a comparison of native languages such as C/C++ and managed runtime environments such as .NET. The project will also evaluate a number of different APIs and libraries that have been developed to support multi-threaded programming. The research will conducted primarily on the MQUTeR cluster computer whose nodes consists of dual quad core (Xeon E5310 @ 1.6Ghz) processors with 8GB RAM. Research Centre: MQUTeR Supervisor: Dr Wayne Kelly
Development of Text Mining System for Mining for Customer Opinions Abstract: With the rising and developing of Web2.0 websites, increasing number of websites encourage visitors to post reviews regarding products or services. These reviews are helpful for both information publishers and readers. This sort of feedback could be used to evaluate customer satisfaction about products or services and take appropriate and sensible actions based on this information. Moreover, people could independently evaluate a product by viewing other people’s opinions on that product, which will probably affect their decisions on whether to obtain the product. As a result, the number of reviews that a product receives grows rapidly. Some fashionable products can get hundreds of reviews at some large commercial sites. On the other hand, many reviews are long with few sentences conveying the owners’ opinions. Therefore, it is not easy task for people to find or gather useful information they desire. Moreover, for each information unit to be reviewed, such as a product, there may be many reviews. If only few reviews are read, the opinion will be biased. As a result, automatic review mining and summarization has become useful tools for analysing reviewers’ opinions recently. This project is to develop a Java or C# program to implement a text mining system to extract customer opinions from online customer reviews. Supervisor: Dr Yue Xu
Semantic Annotations to Discovered Knowledge Abstract: Currently there are two main research directions in data mining. One direction is concerned with the design of efficient algorithms for discovering knowledge, and another is concerned with the interpretations of the discovered knowledge.Most research in data mining has so far focused on the former direction and various techniques have been developed for efficiently mining frequent patterns and association rules, but little achievement has been made on interpreting the discovered knowledge due to two obstacles: the overwhelmingly large volume of discovered patterns (or rules) and the lack of semantic information along with the discovered knowledge. The unintelligibility of the discovered knowledge significantly obstructs the effective use of the knowledge and as thus the worth and importance of automatic knowledge discovery depreciates a lot. In many cases, with some specific information and requirements users want to allocate a small set of rules that particularly address the specific problems at hands. Thus, before exploring the discovered knowledge, it would be very helpful and beneficial to provide users with some semantic information about the hidden meaning of the discovered knowledge. Therefore, the goal of this research is to develop techniques to annotate the discovered patterns and associations with semantically enriched descriptions and context information. This semantic information can suggest the potential meaning of the patterns or rules and to help users decide whether and how to explore and use the discovered knowledge. This project will develop a Java or C# program to implement the techniques. Supervisor: Dr Yue Xu
|
||||||||||||||||||||||||||||||
