Research Seminars, Lectures and Events 2007-2008
Seminar: Centre for Human Computer Interaction Design Seminar "Experience, Heuristics, and Information Flow in Requirements Engineering"
Seminar: Department of Information Science Seminar "Search interfaces: how simple can we make them?"
Seminar: Department of Computing (AIS Group) / Department of Psychology Seminar "Neural-Symbolic Systems"
Seminar: Department of Computing (Software Engineering Group) Seminar "Learning and Reasoning in Commonsense Domains"
Seminar: Department of Computing (AIS Group) / Department of Psychology Seminar "How people categorize and reason about the world using prototypes."
Seminar: Department of Computing (PLAS Group) Seminar "Tight Enforcement of Flexible Information-Release Policies for Dynamic Languages ."
Centre for Human Computer Interaction Design Seminar "Experience, Heuristics, and Information Flow in Requirements Engineering"
Speaker: Kurt Schneider, Leibniz Unviersität Hannover, Germany
Location: A222
Date: 28 March 2008
Time: 1pm
Abstract
A technique for improving the information flow in requirements engineering will be presented: It helps to speed up elicitation and validation in a large project. The technique supports interviews with a mobile tailor-made tool with built-in use case animation features. A requirements analyst can elicit requirements and validate them immediately by looking ahead to sketches of user interfaces. A tablet-PC is used as a medium to facilitate fast feedback in requirements analysis interviews. This technique is part of an approach to improve information flow in requirements engineering by heuristic and experience-based techniques and tools.Profile
Kurt Schneider studied Computer Science in Erlangen, Germany. He received a Ph.D. in Software Engineering from the University of Stuttgart in 1994. He had a grant from the NATO Science Committee for a Postdoctoral position at the University of Colorado at Boulder. Kurt Schneider was a visiting member of the interdisciplinary Center for LifeLong Learning and Design in Boulder. From 1996-2003, he held a position at the DaimlerChrysler Research Center in Ulm, Germany. In 2003, he became full professor of Software Engineering at Leibniz Universität Hannover. His main research interests are requirements engineering, software quality, and service-oriented architectures.
Department of Information Science Seminar "Search interfaces: how simple can we make them?"
Speaker: Tamar Sadeh
Location: Level 3C, University Building, City University London
Date: 4 April 2008
Time: 12 noon - 1pm
Abstract
Information-seeking in the scholarly environment has undergone major changes in the last few years as searching for information in general has taken on an increasingly central role in our everyday activities. Researchers now expect their search for scholarly information to be as simple and straightforward as their search for other types of information; as a result, many researchers are drawn to information resources that provide simple interfaces but do not necessarily offer the most valuable materials. Providers of scholarly materials, who must deal with large, rapidly growing amounts of data, are looking for ways to accommodate researchers' expectations and still enable researchers to find the exact information they need.
In this seminar, I will introduce a research project aimed at understanding the behaviour and preferences of today's researchers; the outcome of the project will be the proposal of a model for a search interface that is likely to fulfil researchers' requirements and that can be feasibly implemented by providers of information resources and developers of library-related software.
After outlining the status of search interfaces in the scholarly environment, I will present the research hypothesis, an overview of the literature, the aims and objectives of the research, and the methodology that I have decided to use. I will then summarize the research activities that I have carried out so far, discuss the conclusions that I have drawn from those activities, and describe the next steps in the project.
Profile
Tamar Sadeh is a research student in the Department of Information Science supervised by Dr Andrew MacFarlane and Professor Stephen Robertson
Department of Computing (AIS Group) / Department of Psychology Seminar "Neural-Symbolic Systems"
Speaker: Dr. Artur Garcez
Location: Regretably this event has now been cancelled
Date: Regretably this event has now been cancelled.
Time: Regretably this event has now been cancelled
Abstract
Three notable hallmarks of intelligent cognition are the ability to draw rational conclusions, the ability to make plausible assumptions, and the ability to generalise from experience. Although human cognition often involves the interaction of these three abilities, in artificial intelligence they are typically studied in isolation. In our research programme, we seek to integrate the three abilities within neural computation, offering a unified framework for learning and reasoning that exploits the parallelism and robustness of connectionism. A neural network can be the machine for computation, inductive learning, and effective reasoning, while logic provides rigour, modularity, and explanation capability to the network. We call such systems, combining a connectionist learning component with a logical reasoning component, "neural-symbolic learning systems". In this talk, I review the work on neural-symbolic learning systems, starting with logic programming, which has already provided contributions to problems in bioinformatics and engineering. I then look at how to represent modal logic and other forms of non-classical reasoning in neural networks. The model consists of a network ensemble, each network representing the knowledge of an agent (or possible world) in a particular time-point. Ensembles may be seen as in different levels of abstraction so that networks may be fibred onto (combined with) other networks to form a modular structure combining different logical systems or, for example, object-level and meta-level knowledge. Networks may also be combined to represent (and learn) relations between objects, with interesting applications in graph mining and link analysis in biology and social networks. We claim that this quite powerful yet simple structure offers a basis for an expressive yet computationally tractable cognitive model of integrated reasoning and robust learning. The material is part of the forthcoming book "neural-symbolic cognitive reasoning".
This is the first in a series of joint seminars between the Department of Computing and the The Department of Psychology
Profile
Artur Garcez is a Reader at the Department of Computing, School of Informatics, City University London. He is an author of the book Neural-Symbolic Learning Systems: Foundations and Applications, Springer-Verlag, 2002. He is area editor (Reasoning and Learning) of the Journal of Logic and Computation (OUP), editor (Logic and Neural Networks) of the Journal of Applied Logic (Elsevier), member of the editorial board of the International Journal of Hybrid Intelligent Systems (IOS Press), associate editor of the International Journal on Artificial Intelligence Tools (World Scientific), associate member of Behavioral and Brain Sciences (CUP), and member of the advisory board of the Cognitive Technologies book series (Springer-Verlag). Artur Garcez holds an M.Eng. in Computing Engineering, an M.Sc. in Computing and Systems Engineering and a Ph.D. in Computing (D.I.C). He holds a Visiting Research Fellowship at the Department of Computer Science, King's College London. He is a member of the City and Guilds College Association, a member of the British Computer Society, and Research Consultant for Performance Sistemas & Metodos, Ltd.
Department of Computing (Software Engineering Group in conjunction with the SERENITY research project) Seminar "Learning and Reasoning in Commonsense Domains"
Speaker: Dr.Oliver Ray
Location: Regretably this event has now been cancelled
Date: Regretably this event has now been cancelled
Time: Regretably this event has now been cancelled
Abstract
This talk will demonstrate how a nonmonotonic logical reasoning system called XHAIL (eXtended Hybrid Abductive Inductive Learning) can be used to integrate inductive learning and abductive reasoning in two commonsense domains. The first example concerns a biological model for analysing so-called synthetic lethal mutations; and the second involves a general purpose temporal event calculus for specifying state-based systems and processes. I will show how XHAIL automatically solves a range of projection, planning and learning problems according to the prescribed language bias.
Profile
Dr. Oliver Ray is a Research Fellow with The Department of Computer Science at The University of Bristol.
Department of Computing (AIS Group) / Department of Psychology Seminar "How people categorize and reason about the world using prototypes. "
Speaker: Prof. James A. Hampton
Location: E138, City University
Date: 14 May 2008
Time: 2pm - 3pm
Abstract
Prof. Hampton will review some of his recent studies on prototypes in human categorization and reasoning. The general methodology is to give students tasks in which they have to make judgments about how they categorize the world, or whether certain statements about general knowledge are true or false. Issues addressed include the following: Do we have meta-awareness of vagueness when judging whether statements are true or false? How do people combine concepts into compounds such as Pet Bird or conjunctive phrases such as Sports which are Games? How does your belief in a generic statement (ravens are black) change when a modifier is added so that it refers to an atypical subset (jungle ravens are black)?
This is the second in a series of joint seminars between the Department of Computing and the The Department of Psychology
Profile
Professor James Hampton read Natural Sciences at Cambridge, before completing his doctoral degree in Psychology at University College London in 1976. After a year's post-doctoral fellowship in Social Anthropology at UCL with Mary Douglas, he joined City University in 1977. He is currently Professor in Psychology, and was Head of the Department of Psychology from its creation as an independent department in the School of Human and Social Sciences in 1993 until 1998.
He has also held visiting professorships at Stanford (1984-5) and Cornell (1988) Universities in the USA, and at the CNRS in Paris (1995), and in 1996 was visiting professor at the University of Chicago as holder of a Fullbright Senior Fellowship. In 2003 he was Visiting Professor in psychology at Yale University and was appointed a Fellow of Morse College, Yale.
His research interests are in the cognitive processes involved in categorization and in the mental representation of knowledge in conceptual structures and how this is related to word meaning. He has consultancy interests in statistical data analysis and database programming.
He is married with two children and lives in Islington. He also has a retreat in rural France.
Department of Computing (PLAS Group) Seminar "Tight Enforcement of Flexible Information-Release Policies for Dynamic Languages . "
Speaker: Prof. Andrei Sabelfeld
Location: A227, College Building, City University
Date: 20 May 2008
Time: 2pm - 3pm
Abstract
Information-flow tracking in web applications provides a viable, and increasingly popular, alternative for enforcing end-to-confidentiality and integrity. However, there is an unsettling gap in the state of the art between formal, mostly static, approaches---that lack support for dynamic code evaluation---and practical, mostly dynamic, approaches---that lack soundness and support for flexible information-release policies. Seeking to bridge the gap, this talk proposes (i) an intuitive yet surprisingly general framework for rich information-release policies expressing both *what* can be released by an application and *where* in the code this release may take place and (ii) tight and modular enforcement by hybrid mechanisms that combine monitoring with on-the-fly static analysis for a language with dynamic code evaluation and communication primitives. The policy framework and enforcement mechanisms support both termination-sensitive and insensitive security policies. Joint work with Aslan Askarov.
Profile
Andrei Sabelfeld is an Associate Professor in the Department of Computer Science and Engineering at Chalmers University of Technology in Gothenburg, Sweden. After receiving his Ph.D. in Computer Science from Chalmers in 2001 and before joining Chalmers as faculty in 2004, he was a Research Associate at Cornell University in Ithaca, NY. His research has developed the link between two areas of Computer Science: Programming Languages and Computer Security. Sabelfeld's article on Language-Based Information-Flow Security is one of the most cited articles in all of Computer Science from 2003 (source: citeseer).