http://www.comp.rgu.ac.uk/docs/seminars/ The IDEAS Research Institute regularly runs seminars on selected topics. Those include research and inductrial themes of general interest with a computing context. The seminars are open to members of the Robert Gordon and Aberdeen Universities and visitors alike. If you would like to present a topic yourself or you know someone whom would be interested, please get in touch. Enjoy the seminars. Dr. Daniel C. Doolan. en-uk 300 Mon, 19 Oct 2009 12:00:00 GMT Mon, 27 Jun 2011 12:29:00 GMT Computing Fri, 24 Jun 2011 16:38:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110712/ This talk presents a methodology for learning graphical models with hidden nodes that I have been studying with collaborators in recent years. The idea is to employ algebraic techniques (in particular, matrix decomposition and spectral methods) to learn unobserved quantities in graphical models. The talk focuses on tree models, and covers two aspects of the underlying learning problem: parameter estimation and structural learning. The first part is concerned with parameter estimation, where an algorithm called learnHMM is presented that learns hidden Markov models. It is shown that this method can efficiently recover the correct HMM dynamics with a sample complexity depending on some mild conditions of the underlying system. The advantage of this approach over some traditional methods (such as EM) is that our algorithm does not suffer from local minimum issues in nonconvex optimization, and it handles high dimensional observations and long range dependencies more easily. The method can be extended to estimating parameters for nonlinear systems and general tree structured graphical models with unobserved nodes. The second part is concerned with structural learning, where an algorithm is presented to learn the underlying tree topology of a broad class of multivariate tree models with hidden nodes. Exact recovery of the tree structure can be established based on certain natural dependencies on statistical and structural properties of the underlying joint distribution. This method handles high dimensional observations and is more general than existing approaches. Collaborators: Daniel Hsu, Sham Kakade, Anima Anandkumar, Le Song http://www.comp.rgu.ac.uk/docs/seminars/documents/20110712/ Computing Fri, 24 Jun 2011 13:38:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110707/ In this paper we present a novel hybrid representation for Music Information Retrieval. Our representation is built by incorporating audio content into the tag space in a tag-track matrix, and then learning hybrid concepts using latent semantic analysis. We apply this representation to the task of music recommendation, using similarity-based retrieval from a query music track. We are interested in measuring the recommendation quality, and the rate at which cold-start tracks are recommended. We develop a new approach to evaluating music recommender systems, which is based upon the relationship of users liking tracks. Our hybrid representation is able to outperform a tag-only representation, in terms of both recommendation quality and the rate that cold-start tracks are introduced to the system. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110707/ Computing Fri, 27 May 2011 13:38:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110603/ I describe a mathematical systems modelling framework that is motivated by a desire to represent and reason about properties of (large-scale) systems situated in dynamic environments. Motivated by the concepts of distributed systems theory, the framework has at its core mathematical treatments of environment, location, resource, and process, and comes along with a separating modal logic. Extensions to analyze questions in computer security are also considered. The mathematical structures provide a semantics for a modelling tool, called (Core) Gnosis, that, together with some elementary utility theory, has been deployed in a range of commercial projects undertaken with Hewlett-Packards information security business and its customers. I conclude by discussing the role of economics in the context of modelling questions in information security. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110603/ Computing Fri, 27 May 2011 13:33:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110601/ The problem: Electronic patient records contain a mixture of coded information and free text. Information is variably recorded by practitioners as coded or free text, but free text is difficult to convert into forms suitable for quantitative analysis. Important information for health services research or epidemiology may be "hidden" to an unknown extent in free text. In the PREP project we aim to develop generalisable methods for the identification and interrogation of potentially important data "concealed" in free text, use the results to enhance coded data, and evaluate the utility of this approach. The study integrates expertise in computational linguistics, user centred social informatics, statistical/epidemiological and visualization, in a multidisciplinary study. Our methods: Natural Language Processing (NLP) be used to search the free text of large quantities of anonymised free text patient records, and to enhance coded data with pseudo-codes. Statistical methods will be used to explore the impact of integrating the additional information on (a) prevalence estimates (rheumatoid arthritis), and (b) estimates of dates of first relevant presentation (ovarian cancer). A visualization tool for the integrated graphical display of coded and NLP generated data is being developed. It will be used to validate the novel data through clinician and researcher review, and thus to explore the value of these techniques in improving the quality and accessibility of information in electronic patient records. Through user centred social informatics approaches, we are also explore what influences clinicians in the balance between recording free text vs using standard codes (e.g 002.23 Appendicectomy), and how information needs to be stored for it to be useful to and retrievable by clinicians. Our collaboration: This Wellcome Trust funded study is a collaboration between Brighton and Sussex Medical School, the Universities of Sussex and Brighton and University College London. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110601/ Computing Fri, 20 May 2011 13:33:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110527/ Self-organised criticality is a principle for the generation of complexity in nature. It is characterised by power-law event distributions and has been used to describe phenomena in domains such as the dynamics of neural activity, natural evolution and biological motor control. We show that critical behaviour is brought about in a natural way in neural networks and can be achieved in autonomous robots by the optimisation of conflicting goals. The critical dynamics in these systems leads to self-organisation of behavioural options that later can be composed into meaningful behaviours e.g. by reinforcement learning. Furthermore, we discuss applications of the approach in prosthetics as well as methods for guidance of the process of behavioural self-organisation in order to bias the emerging behaviour towards promising regions in the behavioural space, to include background knowledge or to achieve more coherent representations. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110527/ Computing Fri, 20 May 2011 13:31:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110524/ The presentation will cover several of the research topics and corresponding national and international projects ongoing at the Department of Informatics from the West University of Timisoara, Romania. These topics are mainly related to grid and cloud computing, high performance computing, multi-agent systems, nature inspired metaheuristics and computational mathematics. It will be also shortly presented the computational infrastructure which currently consists of a 400 cores cluster and a BlueGene/P. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110524/ Computing Fri, 13 May 2011 13:31:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110520/ Probabilistic modelling is central to modern statistical machine learning, because it allows us to connect prior knowledge to data and uncertainty. New applications of probabilistic modelling have the potential to not only be useful in their own right, but also to motivate new methodological and theoretical advances. In this talk I will discuss the problem of predicting the performance of large, distributed computer systems, motivated by distributed applications such as the data centre applications of Google, Yahoo!, and Amazon. I will describe how these applications raise interesting new problems for machine learning, and how a probabilistic modelling perspective is opening up new techniques for debugging these systems when their performance is poor. I will aim for the talk to be relevant not only to researchers in artificial intelligence and machine learning, but also to researchers in performance modelling and autonomic computing. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110520/ Computing Fri, 06 May 2011 13:31:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110513/ In this talk I will present some of the work being done in the new Inference, Dynamics, and Interaction group, at the University of Glasgow. In particular, we are interested in using probabilistic inference to improve interaction technology on handheld devices (particularly with touch screens). I will show how we are using sequential Monte-Carlo techniques to infer distributions over user inputs which can be (1) augmented with applications to provide a smooth handover of control between the human and device and (2) used to extract additional information regarding touch interactions and subsequently improve touch accuracy. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110513/ Computing Fri, 29 Apr 2011 13:31:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110509/ Although peer review is a normal part of both academic and professional practice, undergraduates are rarely given the opportunity to develop this skill. This is puzzling, as the effort required to incorporate a peer review activity into a course can be minimal and the benefits numerous. In this talk, I will present the Aropa web-based peer review tool, and discuss how it has been used in courses at Glasgow and Auckland. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110509/ Computing Fri, 29 Apr 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110506/ Abstract: Mobile phones are increasingly equipped with sensors, such as accelerometers, GPS receivers, proximity sensors and cameras, that can be used to sense and interpret people behaviour in real-time. Novel user-centered sensing applications can be built by exploiting the availability of such technologies in these devices that are part of our everyday experience. Moreover, data extracted from the sensors can also be used to model people behaviour and movement patterns providing a very rich set of multi-dimensional data, which can be extremely useful for social science, marketing and epidemiological studies. In this talk I will present some of my recent work in this area including the design and implementation of the CenceMe platform, a system that supports the inference of activities and other presence information of individuals using off-the-shelf sensor-enabled phones and of EmotionSense, a system for supporting social psychology research. Finally, I will briefly discuss how we are using these data to model people activity and movement patterns in the geographical space. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110506/ Computing Fri, 08 Apr 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110415/ Institutional theory and the Capabilities Approach have become influential in international development research and practice. Both theories offer analytical tools for interpreting and guiding Information and Communication Technology (ICT) for adoption interventions. An analytical framework is proposed which applies Institutional Theory and the Capabilities Approach in the domain of ICT enabled technology change. It is argued, using empirical evidence from a case study in sub-Saharan Africa, that there are benefits in research and practice of utilizing the contrasting strengths of these analytical tools. A combined theoretical framework offers analytical and practical insights in terms of potential for stimulation ("excitation") and degradation ("inhibition") of technological change goals. ICT can create changes in institutional rules and norms enabling enhanced capabilities for individuals. It can also enhance capabilities which in turn lead to institutional change. There are counter-examples, where ICT has negative influences and the lack of capabilities or institutional rules or norms can prevent fulfilment of hoped for benefits from its interventions. This work contributes a combined framework linking both theories and the attendant exciters and inhibitors, illustrated with the case study. The framework contributes to theory development and informing practice by offering a novel approach to analyze ICT led international development interventions. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110415/ Computing Fri, 25 Mar 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110401/ There is more than words in linguistic communication. Whenever involved in social interactions, people display a wide number of nonverbal behavioural cues (facial expressions, vocalisations, gestures, postures, etc.) that add entirely new layers of meaning to the words being uttered. Social Signal Processing is the new, emerging domain aimed at conceptual modelling, automatic analysis and machine synthesis of nonverbal cues used as social signals, i.e. signals conveying information about social actions, social relations, social emotions and social attitudes. The goal of this talk is to illustrate the general aspects of the domain, present some examples of SSP works, and show how SSP can be helpful to make computers more adept and robust to realistic socio-cultural phenomena. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110401/ Computing Fri, 18 Mar 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110318/ The Boolean satisfiability problem, which asks whether a given Boolean formula has a truth assignment of its variables for which the formula evaluates to true, is one of the most fundamental problems in computer science. Apart from being of immense practical relevance, it is intimately connected to the celebrated NP vs P question. In this survey talk, I will motivate the study of this problem, describe some of the known algorithms for it which are known to do better than the trivial brute-force enumeration of truth assignments, and tell you about some exciting recent developments which show that improved algorithms for the Satisfiability problem would lead to new complexity lower bounds. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110318/ Computing Fri, 11 Mar 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110311/ Information visualisation is a research area that focuses on the use of graphical techniques to present data in an explicit form. Such static (pictures) or dynamic presentations help people formulate an understanding of data and an internal model of it for reasoning about. Such pictures of data are an external artefact supporting decision making. While sharing many of the same goals of Scientific Visualisation, Human Computer Interaction, User Interface Design and Computer Graphics, Information Visualisation focuses on the visual presentation of data without a physical or geometric form. As such it relies on research in mathematics, data mining, data structures, algorithms, graph drawing, human-computer interaction, cognitive psychology, semiotics, cartography, interactive graphics, imaging and visual design. In this talk I will present a brief history of social network analysis and visualisation, introduce layout algorithms we have developed for such visualisation and provide a detailed case study on the layout of evolving or "dynamic graphs extracted through our process of "social network inference" from large 10m records without explicit relations. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110311/ Computing Fri, 4 Mar 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110304/ In many important situations, giving consideration to the time and resource impact of planning decisions made is essential to allow useful plans to be produced. For instance, if the plan is to be executed by an agent with limited battery capacity, the planner must choose actions that respect this limit, including actions to replenish the battery if appropriate. Or, if there is a deadline by which certain goals must be met, then it is important that the actions complete within the desired time-scale. In this talk, I'll be presenting a flexible planning architecture, POPF, for problems such as these. Its kernel works by constructing a 'Partial Order Plan, Forwards', allowing it to find temporally efficient plans. In its original form, this was combined with a Linear Programming (LP) solver, for solving problems where actions interact with resources continuously over their execution. More recently, as part of SICSA, it has been combined with a Bayesian Network, to estimate the likelihood that a given sequence of plan steps will meet a deadline. To motivate the work, I'll present a case study for each of these. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110304/ Computing Fri, 25 Feb 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110225/ Traditional approaches to machine intelligence view systems as "assistants" engaged in a dialog, but many contexts demand intelligence on a continuous fine-grained level. The system becomes an extension of the user, like a tool which constantly adapts for every use. Shared control in multimodal interactions will be discussed using a variety of sensors, in both familiar mobile environments and the less familiar realm of EEG-based brain computer interaction. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110225/ Computing Fri, 11 Feb 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110211/ Humans naturally reuse recalled knowledge to solve problems and this includes understanding the context i.e. the information that identifies or characterizes these problems. For problems in complex and dynamic environments, providing effective solutions by operators requires their understanding of the situation of the environment together with the context. Context-aware case-based reasoning (CBR) applications use the context of users to provide solutions to problems. The combination of context-aware CBR with general domain knowledge has been shown to improve similarity assessment, solving domain specific problems and problems of uncertain knowledge. Whilst these CBR approaches in context awareness address problems of incomplete data and domain specific problems, future problems that are situation-dependent cannot be anticipated due to lack of the facility to predict the state of the environment. This paper builds on prior work to present an approach that combines situation awareness, context awareness, case-based reasoning, and general domain knowledge in a decision support system. In combining these concepts the architecture of this system provides the capability to handle uncertain knowledge and predict the state of the environment in order to solve specific domain problems. The paper evaluates the concepts through a trial implementation in the flow assurance control domain to predict the formation of hydrate in sub-sea oil and gas pipelines. The results show a clear improvement in both similarity assessment and problem solving prediction. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110211/ Computing Fri, 21 Jan 2011 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20110121/ As a result of their ability to deliver high quality solutions in reasonable time, Meta-heuristics are usually employed as effective methods to solve the complex multi-objective optimization problems. One class of such meta-heuristics is Hierarchic Genetic Strategy (HGS). A Genetic Algorithm variant, HGS differs from other genetic methods in its capability of searching concurrently the solution space. The HGS efficiency is therefore produced by the simultaneous execution of many dependent evolutionary processes. Every single process is then interpreted as the branch in a tree structure and can be defined as a sequence of evolving populations. The overall dependency relation among processes has a restricted number of levels. In this talk we present the theoretical and experimental evaluation of HGS in solving various complex multi-objective optimisation problems in discrete and continuous domains. In particular, the application of the strategy in scheduling the independent tasks in Computational Grids is highlighted. http://www.comp.rgu.ac.uk/docs/seminars/documents/20110121/ Computing Fri, 11 Dec 2009 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20091211/ jCOLIBRI framework has been developed by Dr. Recio-Garcia during the last few years and is currently an established tool in the Case-Based Reasoning (CBR) community. In this talk he will review the functionality offered by jCOLIBRI for developing different types of CBR systems: Textual, Knowledge Intensive, Data Intensive, and Recommender Applications. He will also present his outgoing work at RGU for including web knowledge and resources in the CBR process. http://www.comp.rgu.ac.uk/docs/seminars/documents/20091211/ Computing Fri, 04 Dec 2009 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20091204/ Automated Question Answering (QA) remains both challenging and important task for computational linguists, artificial intelligence and information retrieval researchers. I will present work in progress that attempts to formulate and explore 'active approach' to QA, which may be considered an alternative to a more popular and extensively studied 'passive' approach in which all the necessary resources (such as ontologies, grammars, transformations, patterns and numerous other heuristics) are built first... http://www.comp.rgu.ac.uk/docs/seminars/documents/20091204/ Computing Fri, 13 Nov 2009 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20091113/ Document genre (e.g. scientific article, fiction, advertisement) is closely bound to the physical and conceptual structure of the document as well as the level of depth and reliability of the information found within the text. Hence, it is useful as a reference for comparing documents on the basis of metrics other than topical similarity. Despite its usefulness, the success of previous attempts to automate genre classification is somewhat unsatisfactory. In this talk ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20091113/ Computing Fri, 9 Oct 2009 13:30:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20091030/ Linkage learning has been a focus of research interest since the early days of evolutionary computation. There is a strong connection between linkage learning and the concept of structure learning, which is a crucial component of a multivariate Estimation of Distribution Algorithm. Structure learning determines the interactions between variables in the probabilistic model of an EDA, based on analysis of the fitness function or a population.. http://www.comp.rgu.ac.uk/docs/seminars/documents/20091030/ Computing Fri, 23 Oct 2009 13:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20091023/ Modelling and recognition of human faces are interrelated disciplines. The challenge in developing a facial model is not only to have a model that looks realistic but one which also can be efficiently utilized for the specific purpose for which the geometry is created. For recognition purposes, the processing of 3D facial data is an essential step, because these images may have outliers, or noise, and could be in any arbitrary orientation. Hence, by processing these images certain facial characteristics could be identified... http://www.comp.rgu.ac.uk/docs/seminars/documents/20091023/ Computing Fri, 02 Oct 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20091002/ This talk will discuss the business need for remote speaker verification solutions and will detail the supervised and unsupervised neural network based methods that are being developed at Nottingham Trent University to address this need. The talk will include an overview of a practical VoIP based computer network biometric logon demonstrator that has been produced and will conclude with a short account of the spin-out company creation processes that are currently being undertaken to commercially exploit the technology. http://www.comp.rgu.ac.uk/docs/seminars/documents/20091002/ Computing Thu, 24 Sep 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20090924/ Many optimization problems in the real world, by nature, have multiple conflicting objectives. Unlike a single optimization problem, a multiobjective optimization problem (MOP) has a set of Pareto optimal solutions (Pareto front) which could be required by a decision maker to make her final decision. Evolutionary algorithms are able to generate an approximation to the Pareto front in a single run, and many traditional optimization methods have been also developed for solving MOPs. Although there is not much work that has been done ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20090924/ Computing Wed, 17 Jun 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20090617/ In many decision-making contexts, choice is taken not by isolated individuals but is rather the result of a process of interaction among many incumbents. "Deliberative Agreement" intends to explore and analyze those situations where a decision is made by a group of individuals in the context of multiagent systems and distributed computing. The overall goal of our ongoing research is to analyze properties and develop mechanisms for collective decision making in human and artifcial agents. The focus will be on tasks that require complex agreements, i.e. involving at least one of the following: (i) argumentation processes for deliberation, (ii) aggregation of sets of interconnected judgments. http://www.comp.rgu.ac.uk/docs/seminars/documents/20090617/ Computing Mon, 01 Jun 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20090601/ The openness and anonymity of the Internet environment create many hazards for e-commerce. For collaborative recommender systems, it raises the possibility of that attackers will seek to bias the output recommendations through manipulation of the public inputs that the system permits. Fighting such manipulation is a constant battle for the owners and maintainers of such systems. In this talk, ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20090601/ Computing Fri, 20 Mar 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20090320/ Security requirements for a software system often change. However, imprecise traceability links discovered or documented between requirements and implementations can make the security analysis difficult. Therefore it is important for the analyst to verify secure requirements on the implementation and to reflect the impact of changes on both sides ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20090320/ Computing Fri, 13 Mar 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20090313/ We propose textual reuse as the identification of reusable textual constructs in a retrieved solution text. This is done by annotating a solution text so that reusable sections are identifiable from those that need revision. We present a novel and generic architecture, Case Retrieval Reuse Net (CR2N), that can be used to generate these annotations to denote text content as reusable or not ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20090313/ Computing Fri, 27 Feb 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20090227/ A Distributed Constraint Satisfaction Problem (DisCSP) is a constraint satisfaction problem distributed between several agents who cooperate in order to solve the overall problem. Frequently, these DisCSPs are coarse-grained where each agent has its own subproblem to solve which is linked to other agents' subproblems through inter-agent constraints. Existing techniques to solve DisCSPs can be mainly categorised into backtracking and local search algorithms. Backtracking algorithms are ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20090227/ Computing Fri, 20 Feb 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20090220/ Dr Laura Muir is a Lecturer in the Department of Information Management at Aberdeen Business School and a member of the research team in the Centre for Video Communications (School of Engineering). Her research seminar explores the visual perception mechanisms that influence how video information is communicated and used. She presents the methods and models developed in her PhD research (Content-Prioritised Video Coding for British Sign Language Communication) and how they are being applied to address a new research problem. http://www.comp.rgu.ac.uk/docs/seminars/documents/20090220/ Computing Fri, 13 Feb 2009 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20090213/ Social software technologies have led to an increase in user content on the Internet. In particular users are now able to post their sentiments on products and services with ease on a regular basis. Such postings range from one-line-comments to detailed reviews on movies, books, photo albums, holiday bargains, medical treatment and other experiences that the authors have been through. ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20090213/ Computing Fri, 21 Nov 2008 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20081121/ This is a great opportunity to learn about some brand new technology which is being used in industry and you are likely to be exposed to when you complete your studies. Its also a chance to learn about some of the programmes that Microsoft offers to students which can help you get hold of their software for free and enhance your CV. http://www.comp.rgu.ac.uk/docs/seminars/documents/20081121/ Computing Fri, 21 Nov 2008 11:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20081121b/ Genetic Algorithms (GAs) are a commonly used search heuristic. They are one of a range of Evolutionary Algorithms, i.e. search techniques which follow evolutionary inspired procedures to evolve solutions to problems. GAs have been successfully applied to a plethora of application areas. One area where GAs have been effective is that of deriving schedules for optimal control problems. This talk reviews the work undertaken over the last 3 years of my PhD... http://www.comp.rgu.ac.uk/docs/seminars/documents/20081121b/ Computing Fri, 14 Nov 2008 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20081114/ The facial expression clone project involves modelling facial models with expressions for particular individuals. Creating a facial model from scratch is a labour intensive work, which involves several modelling pipelines; and crafting natural life-like facial expressions is really an artistic endeavour... http://www.comp.rgu.ac.uk/docs/seminars/documents/20081114/ Computing Fri, 07 Nov 2008 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20081107/ Human beings are adept and drawing context-sensitive associations and inferences across a broad range of situations ranging from the mundane to the creative inferences that lead to scientific discovery. Such reasoning has a strongly pragmatic character and is transacted with comparatively scarce cognitive assets. The question is ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20081107/ Computing Fri, 31 Oct 2008 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20081031/ Sharing of experience has always contributed to the collective knowledge of societies. Traditionally experience has been shared in the form of fables and stories passed on from person to person. In well defined domains the role of apprenticeship has been instrumental in knowledge sharing. In the information age computers have started playing a role as facilitators in knowledge sharing. We look at an approach to knowledge management (KM). http://www.comp.rgu.ac.uk/docs/seminars/documents/20081031/ Computing Fri, 24 Oct 2008 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20081024/ Several EDAs based on Markov networks have been recently proposed. Key ideas behind these EDAs were to factorise the joint probability distribution of the solution variables in terms of the clique in the undirected graph, and sample from them to generate the new population. As such, they made use of the global Markov property of the Markov network. In this talk, I will present a Markov Network based EDA ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20081024/ Computing Fri, 17 Oct 2008 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20081017/ The first decade of the 21st century has seen tremendous uptake in the use of mobile phones. Less than a year ago it was announced that 50% of the worlds population had mobile phone subscriptions. It is expected that this figure will rise significantly by the end of the decade. Mobiles are generally used for voice communications and text messaging, but they are capable of much more. The world of today is ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20081017/ Computing Fri, 10 Oct 2008 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20081010/ This research focuses on overlaps of information retrieval (IR), cognitive science and genre, merging and utilizing these for one particular goal: to analyse and retrieve structured text. Structured textual documents are normally composed of several layers or sections which together form types, or genres, of text preserved, in particular, in e-mail and Wikipedia (XHTML)... http://www.comp.rgu.ac.uk/docs/seminars/documents/20081010/ Computing Fri, 16 May 2008 14:00:00 GMT http://www.comp.rgu.ac.uk/docs/seminars/documents/20080516/ Selection is one of the defining characteristics of an evolutionary algorithm, yet inherent in the selection process is the loss of some information from a population. Poor solutions may provide information about how to bias the search toward good solutions. Many Estimation of Distribution Algorithms (EDAs) use truncation selection which discards all solutions below a certain fitness, thus losing this information ... http://www.comp.rgu.ac.uk/docs/seminars/documents/20080516/