Artificial Intelligence Techniques in Software Engineering

Masoud Mohammadian

Artificial Intelligence (AI) techniques have been applied in many areas of software engineering. However due to the complexity of software systems these techniques are usually considered in research laboratories and their use in industry is very limited. This talk provides an insight onto applications fo AI techniques in software engineering and how AI can assist in achieving ever competitive and firm schedules for software development projects. Some pros and cons of using AI techniques are investigated and specifically the application of AI in software security assessment and data security is considered.

This talk also will introduce the application on an AI technique namely Fuzzy Cognitive Maps (FCM) for risk assessment in software system design. I demostrate how FCM can be adapted in software system design to provide facilities that will enable the system architects to perform what-if analysis and to better understand vulnerabilities of their designed system.

July 2009

Reflection on the Experience gained from undertaking my OSP in the School of Electronics and Computer Science at KingsCollege London

Dr Bala Balachandran

The aim of this seminar is to share and inform about my experience gained from undertaking my OSP program in the School of Electronics and Computer Science at Kings college London. The seminar will overview the courses, curriculum and teaching methods used in the School of Electronics and Computer Science at kings College. Based on the experience I will outline some of my current activities at ISE in terms of new unit design and teaching. The seminar will also discuss the research work on the area of automated negotioation that I carried out at KCL as part of my OSP program.

Correlation, regression and "transmuting" women into men: a celebration of the life of Francis Galton

Dr Alice Richardson

This year the scientific community is celebrating the 200th birthday of Charles Darwin, who through his book "The Origin of Species" contributed enormously to our understanding of evolution and the natural world. Charles Darwin's extended family was extraordinarily talented, and in this talk I will describe the life and achievements of his cousin, statistician Francis Galton. Galton invented the correlation coefficient, introduced the world to regression and even contributed to early research into the forensic use of fingerprints. I will describe the context of each of these innovation, and show how their usefullness has been extended in the century since Galton introduced them.

May 2009

Plasmas in Science and Engineering

Professor John Rayner

Plasma is the most common form of visible matter in the Universe. It is the stuff of galaxies, stars and a range of natural phenomena. It is also an incredibly important engineering material with applications ranging from fusion as a primary energy source through to coatings for artificial hips and stealthy antennas. In this talk I will look at some of these ideas and also describe the work that has been going on in the plasma Instumentation laboratory in the Faculty on wave-heated plasmas and plasma antennas.

May 2009

Reflection on the Experienced gained from undertaking my OSP in the School of Electronics and Computer Science at King's College London

Dr Bala Balachandran

The aim of this seminar is to share and inform about my experience gained from undertaking my OSP program in the School of Electronics and Computer Science at King's College London. The seminar will overview the coursed, curriculum and teaching methods used in the School of Electronics and Computer Science at King's College. Based on the experience I will outline some of my current activities at ISE in terms of new unit design and teaching. The seminar will also discuss the research work in the area of automated negotiation that I carried out at KCL as part of my OSP program.

May 2009

Industry/Academic Collaboration

Kevin Cox

This seminar will introduce/reintroduce Kevin Cox as a new adjunct member of staff. It will describe how Edentiti developed its business model and how the software and software methodologies supported the business mode. Kevin will outline his and the Companies current interests and it will ask attendees to bring ideas on how they see industry academic collaboration working and how they think they can personally use an organisation like ours to assist their research, teaching, or organisational activities. Edentiti is representative of the many small innovative companies in Canberra and can provide introductions and links to other people in this sector of the ACT economy.

Edentiti the company is in the business of providing individuals with an electronic 'presence' that they can use to help them with their online activities and their interactions with organisations. The talk will describe how we have translated the idea into a business proposition and the directions the company is likely to evolve. The seminar will end with an open discussion on ideas to involve both Kevin Cox, his contacts and Edentiti in attendees activities and interests

May 2009

Enhanced laboratory Diagnosis of Human Chlamydia pneumoniae Infection through Pattern Recognition Derived from Pathology Database Analysis

Dr Alice Richardson/Dr Brett Lidbury

This study focuses on pattern recognition in pathology data collected from patients tested for Chlamydia pneumoniae (CP) infection, with co-infection by Mycoplasm pneumoniae (Myco) also considered. Both Cp and Myco are microbes that cause respiratory disease in some infected people. As well as the immunoassay results revealing whether the patient had been infected, or not, an extensive range of other routine pathology data was also available for each patient, allowing the analysis of associations between a positive immunoassay laboratory result for Cp or Myco, and a range of tests for biochemical and cellular markers.(e.g. liver enzymes, electrolyte balance, haematological indices such as red/white cell counts). Decision trees and logistic regression were used to enhance laboratory diagnosis of these respiratory infections via the formulation of association rules derived from immunoassay results and associated pathology data.

May 2009

Negative Selection with Antigen Feedback in Intrusion Detection

Dr Wanli Ma

One of the major challenges for negative selection is to efficiently generate effective detectors. The experiment in the past shows that random generation fails to generate useful detectors within acceptable time duration. In this paper, we propose an antigen feedback mechanism for generating the detectors. For an unmatched antigen, we make a copy of the antigen and trat it the same as a newly randomly generated antibody: it goes through the same maturing process and is subject to elimination due to self matching. If it survives and is then activated by more antigens, it becomes a legitimate detector. Our experiment demonstrates that the antigen feedback mechanism provides an efficient way to generate enough effective detectors within a very short period of time. With the antigen feedback mechanism, we achieved 95.21% detection rate on attack strings, with 4.79% false negative rate, and 99.21% detection rate on normal strings, 0.79% false positive. In this paper, we also introduce Arisytis - Artificial Immune System Tool Kits - a project we are undertaking for not only our own experiment, but also the research communities in the same area to avoid the waste on repeatedly developing similar software. Arisytis is available on the public domain. Finally, we also discuss the effectiveness of the r-continuous bits match and its impact on data presentation.

March 2009

What data should be encrypted?

Masoud Mohammadian

Organizations understand the value of their data and need for adequate data protction services. Data security and privacy are very important issues in the success of a business operation. Implementing and applying policies related to data security and privacy therefore has become one of the core and important activity for large organizations. Data classification process allows companies to organize their information according to their needs. This process can be a laborious in large organizations with significant content to evaluate and categorize. Using a data classification process organizations can identify and encrypt only the required data based on orgaisational policies thereby saving time and processing power. This paper explores the use of fuzzy logic in classification of dat and suggests a method that can determine requirements for data security and privacy in an organization based on organizations needs and government policies imposed on data.

March 2009