phd research topics - robert gordon · pdf filephd research topics ... etc. the increasing...

13
PhD Research Topics SCHOOL OF COMPUTING SCIENCE AND DIGITAL MEDIA

Upload: lamlien

Post on 19-Mar-2018

214 views

Category:

Documents


1 download

TRANSCRIPT

PhD Research Topics

SCHOOL OF COMPUTING SCIENCE AND DIGITAL MEDIA

foreword:The School of Computing Science and Digital Media has internationally recognised research pushing the boundaries of theoretical and practical knowledge in Computing and offering novel thinking on how technology can enrich society and enable business to perform at a higher level. The School’s research focuses on tackling challenges related to Big Data and Internet of Things. The main research themes include:

• Machine Learning and Optimisation• Information Retrieval• Artificial Intelligence• Data Visualisation• Usability

Researchers from the School actively pursue project opportunities and collaborations related to the the strategic priority themes for the university of Oil and Gas, Smart Data, Remote Healthcare and Northern Culture and Design.

The Smart Data Technologies Centre (SDTC) is co-located with the School. The SDTC centre focuses on big data analytics, particularly for the oil and gas sector and aims to secure efficiencies from the use of analytics.

The School of Computing Science and Digital Media works closely with the DataLab – a Scottish innovation centre, which aims to generate significant economic, social and scientific value from big data. Robert Gordon University hosts one of the DataLab hubs, focusing on Oil and Gas.

The School is also a member of the world-leading Scottish Informatics and Computer Science Alliance (SICSA).

desirable experienceCONTACT POINT

TOPIC DETAILS:

Behaviour-based Malware Detection using Machine Learning Techniques

The last decade has seen an important growth in the number of malware propagated over Internet. Anti-malware provides some protection against well-known malware, as well as slight variations of them. It mainly uses signature-based techniques where a database of well-known malware is maintained. However, signature-based techniques are easy to evade. Recent works on malware detection have focused on behaviour-based techniques (actions of malware).

We propose to develop a machine learning-based framework for the automatic detection and classification of both existing and new malware using their behaviour. The framework will allow both clustering (grouping malware with similar characteristics), and classifi-cation (assigning new malware to one or more classes).

Our proposal addresses the following research questions:1. Can machine learning-based malware detection be effective? This could be measured by metrics such as false-positives and false-negatives.2. Can behaviour-based techniques be efficient? As they are known to be slower than signa-ture-based ones, are there mechanisms that can make them fast enough to become inter-esting enough.3. Which machine learning techniques would be suitable and how to optimally use them?4. What are the best criteria for malware classification? The classification could use different criteria such as types of malware and its targeted platform.

• knowledge and or experience in the subject• a good knowledge of machine learning• experience / willingness to learn about mal-ware analysis

Research Themenetworking & computer Security

Dr hatem [email protected] +44 01224 262787

desirable experienceCONTACT POINT

TOPIC DETAILS:

Accessible Music Creation

The purpose of this project is to investigate methods that can be used to combine the rehabilitation exercises that are required following an injury with the ability to perform music.

Typical musical instruments require many years of skill and expertise in order to master. It is hoped that this project will develop more atypical ‘instruments’ that can allow non-musicians to create music with relative ease.

This project, firmly based in the HCI/accessibility domain will require developing an understanding of the exercises needed in rehabilitation and use these exercise (i.e. inputs) as a method for controlling music. A very early example of this can be seen here: https://www.youtube.com/watch?v=sejvNLvJgkk

• Experience or knowledge of creating digital devices• Experience in quantitative and qualitative user testing• A Musical background (minimum of ABRSM Grade 5 Level or equivalent)

Research ThemeUsability

Dr michael [email protected] +44 01224 262783

desirable experienceCONTACT POINT

TOPIC DETAILS:

AI Technologies for Transforming Data into Smart Information Systems

The intersection of people, data and intelligent algorithms has far-reaching impacts. Companies collect vast amounts of data from their customers, processes, and physical infrastructures and people generate lots of data in their digital footprint through their digital interactions.

This research topic is interested in extracting knowledge from collections of data in order to build intelligent retrieve and reuse systems that help companies make decisions and recommendations, and enable people to interact and engage with information. Smart information systems need robust intelligence that incorporates agility and responsiveness to complex, changing scenarios; awareness of individual, social and physical contexts; and long-term reflection of experiences.

The following are sample applications where this research is relevant:

• Big data analytics for monitoring, diagnosis and prediction in complex/challenging environments• Recommender systems for on-line products and services• interactive systems for visitor engagement with tourist locations There are many other topics associated with this leading edge research topic

• Interest in AI, Data Science and/or Big Data • Strong programming skills

Research Themesartifical intelligenceinformation retrievaldata visualisation

prof susan [email protected] +44 01224 262711

desirable experienceCONTACT POINT

TOPIC DETAILS:

Tomorrows World Today: Mobile, Cloud and the Internet of Things

The world of today is quite different from what it was just five years ago. Mobile computing is embedded in every facet of our lives. More and more do we hear about Cloud Computing as apps and services become reliant on this architecture for data synchronisation, storage and analysis.

The world of tomorrow is getting ever closer, it will be an interconnected world, filled with trillions of devices all sensing and communicating with the surrounding environment (Internet of things). How can we tap in to this global network to help improve the lives of those around us? The following are some domains you could explore if you wish to be at the forefront of tomorrows world:

• Health Fitness and Wellbeing• Remote Patient Monitoring• Real-time environment sensing• Smart Cities, Countries and Planet

We Would be happy to discuss other opportunities in this area

• strong programming skills in Mobile development • Knowledge of issues relating to health, activity, aging or domain of choice• Interest in remote sensing, embedded systems, mobile, cloud computing

Research Themesmobile computinginternet of things

Dr daniel c. [email protected] +44 01224 262723

desirable experienceCONTACT POINT

TOPIC DETAILS:

Intelligent Machine Vision

Detecting and identifying objects of interest in images/video footage has attracted great attention in the past three decades from different research areas. This is due to the urgent need for having reliable machine vision systems in several applications domains (security, food industry, fisheries, etc. the increasing availability and usage of CCTV and video cameras, which makes manual interpretation of the resulting large volumes of footage both labour intensive and error-prone. In recent years, major advancements have been achieved in this area. For example computerized face recognition systemssurpassed human beings in controlled environments. However, in real-life scenarios (i.e. uncontrolled environment) accuracy often drops significantly.

There are several interesting topics that are related to this domain. If you are interested in pursing a challenging research topic in one of the following areas, please get in touch:

• Face Detection and Classification • Objects detection and tracking in cluttered scenes • Machine Vision for Augmented Reality

We Would be happy to discuss other opportunities in this area

• knowledge or experience of one of more of the following would be an advantage • 3D Face modelling and animation • Advanced Machine Learning • Deep Learning

Research Themesmachine learningartifical intelligence

Dr Eyad [email protected] Tel: +44 01224 262737

desirable experienceCONTACT POINT

TOPIC DETAILS:

Information retrieval and search

Information retrieval and search has been a key enabler in our ability to find, access, and navigate through the vast amount of content available on the web and large-scale data repositories.

The increasing amount and diversity of information available brings challenges and opportunities in a variety of fields. These include the need for real-time information, increase in type and variety of content that also takes on board multimedia (image retrieval, video retrieval, etc.) and a wider variety of tasks, performed by increasingly diverse end-users.

This research area is interested in developing search algorithms and information systems that meet current users’ needs in a variety of possible domains. We have a long standing track record of research in search algorithms and end-user evaluation.

Below is an indicative list of the topic areas we invite applications from: • Social media search, retrieval, and analytics • Citizen data science, trust and verification algorithms • Multimedia retrieval, image retrieval • Real-time text and multimedia search and retrieval

We would be happy to discuss further topics associated with this leading edge research domain.

• Interest in Information Retrieval, Data Science and/or Big Data, Machine Learning • Strong programming skills • Appreciation of end-users, and experience in interactive systems

Research ThemeInformation retrievalMachine LearningData Analytics

prof ayse [email protected] +44 01224 262475

desirable experienceCONTACT POINT

TOPIC DETAILS:

Socially and Physically Inclusive Gaming

Video games are a hugely important part of the digital economy. Video games are estimated to be a $100bn industry worldwide, and the industry continues to grow in the face of considerable real-life and economic pressures.

However, there are significant demographics which are at least partially excluded from full and active participation in gaming. Sometimes this is as a result of disability. Sometimes it is a consequence of mainstream gaming’s portrayal of sex, gender, race and sexuality. These issues must be addressed on multiple levels, from the sociological to the technical. The aim of this project would be to help design better, more inclusive video games.

This project then encompasses a variety of domains, and applications are invited for projects in the following broad areas: • Accessibility support for modern gaming development platforms • The creation of multi-modal video game interfaces • investigation and deconstruction of online social movements such as Gamergate

• Familiar with the broad landscape of video gaming and its cultural conventions• Technically competent in programming, data mining, or extraction of information from social media contexts• knowledge of storyline, narrative and diversity in games development

Research Themeusability

Dr michael [email protected] +44 01224 262710

desirable experienceCONTACT POINT

TOPIC DETAILS:

Developing Self-Aware Case-Based Systems

Case-based systems employ experiential knowledge to support decision-making and recommendation. They learn by capturing, representing and storing new problem-solving experiences, as additional cases, which can then support future reasoning episodes. While this approach has been successful, the systems operate within a limited learning framework. This project will extend the “learning-by-doing” approach with richer, more “human-like'' learning capabilities.

Building on current research at RGU, new introspective approaches will be developed to provide case-based systems with improved awareness and understanding of their prospective problem-solving ability. Where quality is expected to be low new learning strategies will be triggered; strategies may include simulating human curiosity by employing more proactive techniques, or applying transfer learning to leverage knowledge from related domains.

Potential applications for this research include:

• data exploration using adaptive search techniques• big data analytics for monitoring, diagnosis and prediction in complex environments e.g. medical applications• recommender systems for on-line products and services

• Interest in AI, Data Science and/or Big Data analysis• Strong programming skills

Research Themeartifical intelligence

Dr stewart [email protected] +44 01224 262570

desirable experienceCONTACT POINT

TOPIC DETAILS:

Management and Control of Security

Security of information systems has been a critical issue that has faced organizations since the use of computers. Lately the threats to information systems have grown to enormous proportions and dimensions. One of the major forms of threat is spear phishing termed as advanced persistent threats (APT) which target naïve users of informationsystems to open the system to malware. In this regard it has been statistically proved that insiders pose the biggest threat to information systems which throw a prospective domain to do research on. The following are some domains you could explore if you wish to pursue research in the IS security domain:

• securing the end users from phishers • Training end users in proactive threat detection • The dynamics of insider threats• A taxonomy of APT threats from a multi plane perspective • Auditing and governance of security

We Would be happy to discuss other opportunities in this area

• Networking• knowleged of networking and operating systems including Windows server, Linux - Kali • Knowledge of advanced persistent threats• information systems audit and control • interest in analysing data breaches, exploring log files, pattern matching

Research Themenetworking & computer security

Dr mathew [email protected] +44 01224 262734

desirable experienceCONTACT POINT

TOPIC DETAILS:

Exploring Student Transitions within Higher Education Computing

The experience of transitioning to and starting university is a very individual one, with some undergraduates lacking the cultural capital needed to access teaching and learning within higher education. The literature suggests that the process of student induction is an essential part of their transitioning experience, but it is often a one-off event, rather than an on-going process.

The successful candidate would be responsible for exploring the theories and underlying pedagogy in this area. this would include the gathering, analysiis and visualisation of primary data from schools and FE institutions around Scotland, related specifically to student transitions in computing-related subjects.

Based on the student’s abilities, the project scope may vary. This could include items such as the production of software, tools for visualising data, the creation of workshops and games, an online repository, etc.

• An awareness of statistics, and qualittive/quantitative analysis (although further training may be offered to the righ candidate)• A good knowledge of programming• Experience of (or a willingness to quicklY learn) about data visualisation, experimentaL modelling and user-centred design

Research ThemeComputer science education

Dr angela [email protected] +44 01224 262201

Dr mark [email protected]+44 01224 262768

desirable experienceCONTACT POINT

TOPIC DETAILS:

Cyber Sensor of Network Security

The aim of the project is to develop a virtual cyber security sensor aimed at detecting network traffic anomalies. The proposed sensor will utilise a computational intelligence algorithm capable of building a data-driven model of both known and novel network secu-rity attacks by constructs an inference engine that can recognise abnormal network behaviour. Inferences will be based on data that is extracted directly from the stream of incoming packets using an online classification algorithm. This learning algorithm will be developed to comply with the computational requirements of a low-cost network intrusion detection sensor.

The key challenge in this project is to identify the full set of significant factors affecting the type of network traffic. If a manageable number of such factors are identified, then the computational model can be built, given the acceptable quality of available data, and the developed model gets updated by dynamically tuning its parameters.

• Knowledge of issues relating to data analytics and modelling • Interest in computational intelligence and/or machine learning • knowledge of network security

Research Themenetworking & Computer security

Dr andrei [email protected] +44 01224 262788