Helen Cuenca, MSc Project - Runtime Verification Software
I’m currently working in runtime verification, a method to determine whether a run of a computational system satisfies a given correctness property.
My project consists in participating in the 1st International Competition of Software for Runtime Verification, introduced in 2014 as part of the 14th International Conference on Runtime Verification.
This competition is very important because it’s the first one of its kind and the participants are actively involved in the field of runtime verification.
We expect to do very well in the competition by developing a system for runtime verification based on the Quantified Event Automata (QEA) formalism and implementing multiple optimisation techniques.
My supervisors, Dr David Rydeheard and Giles Reger, are very interested and knowledgeable about runtime verification, so I have learnt from them constantly during the work. We have weekly meetings to check the progress of the project and to discuss next steps, and throughout they have been very supportive and motivating.
Abdulrahman Hussein, MSc Project - Scenario-based requirement modeeling.
The context of this project is set by a recent study which introduced a pattern language to create a set of multi-perspective requirement models from scenarios. Creating these models manually seemed to take time and effort. This project looks into creating an object life cycle model using UML state charts, using the Business Process Model and Notation (BPMN) method. The process model is assumed to be generated from a given scenario at the early stages of requirements elicitation. The project will make use of the Epsilon platform distributed as part of the Eclipse Modelling Project. It is hoped that by automatically creating the target object model, this will help to minimize the efforts required to create requirement models for the purpose of requirements consistency validation or for further software generation.
I would like to thank my supervisor, Dr Liping Zhao for her continuous support through the project up to this time. I have to admit that I was extremely worried at the beginning of the project as this is my first time into the world of model-driven development. With the invaluable guidance and encouragement of my supervisor I managed to overcome the hard times. The project presents a real challenge for me yet my aim is to learn something new and I hope that I will be able to successfully deliver something valuable at the end of the project.
Dommy Asfiandy, MSc Project - Monitoring and Analysis Tools for Mental Health Self-Management
The ESRPC-funded SAMS (Software Architecture for Mental health Self-management) research aims to develop a means to non-intrusively promote self-awareness of change in cognitive function. Such change in older people may indicate dementia problems, particularly Alzheimer’s disease (AD) and Mild Cognitive Impairment (MCI). Those problems generally can only be diagnosed if the patients present themselves for examination by a medical professional. Promoting self-diagnosis should help early detection, so that the treatment and support for people with dementia can be improved significantly.
Fig 1. User activity monitoring tool
My MSc project intends to build monitoring and analysis tools as the underlying framework for SAMS. The tools monitor a set of pre-defined user activities via mouse and keyboard; the captured data are analysed, in collaboration with clinician partners. Then, the results are presented in the form of a dashboard application. Persuasive technology is used to encourage users with potential cognitive problems to seek necessary medical advice.
The discussions with my supervisor regarding the dissertation are noteworthy to aid the development process. At the outset, he helped me to understand the project’s context and continuously provides a substantial guidance. He arranged few meetings with the SAMS’ team members to discuss the key requirements and demonstrate the system to obtain crucial feedback, which would entail a better understanding of the project and developing plausible software.
Aris Kyriakos, MSc Project - Big Data Mining
As datasets across all domains are increasing in size exponentially, industrial and academic organisations are seeking novel ways to process and draw insights from their increasingly voluminous ‘Big Data’ sources. Traditional Data Mining tools, such as Weka and R, fail to cope with these modern data volumes whilst emerging distributed computing frameworks, such as Hadoop and
Spark, can efficiently process petabytes of data, but lack substantial mining libraries. With this in mind, the aim of this project is to combine the plethora of Data Mining algorithms offered by Weka, with the large scale data processing capabilities of Apache Spark. This can be made possible by enclosing Weka’s algorithms in Spark tasks, distribute the tasks to a number of cluster nodes and execute them in parallel on dataset partitions. Since Spark supports in-memory caching mechanisms, one of the primary objectives is to assess the effects of different caching strategies on Big Data Mining workloads.
The project was proposed by Professor John A. Keane and he is also my supervisor. He helps to guide me with weekly meetings and daily emails. It’s really useful to exchange background material, journal articles and conference papers related to the project on a daily basis. In our meetings we also discuss best practices and evaluate progress. My skill-set is constantly evolving and the overall experience during this project will definitely be useful in my future career.