Showing papers by "Nigel Shadbolt published in 2002"

Exploiting synergy between ontologies and recommender systems

Abstract: Recommender systems learn about user preferences over time, automatically finding things of similar interest. This reduces the burden of creating explicit queries. Recommender systems do, however, suffer from cold-start problems where no initial information is available early on upon which to base recommendations. Semantic knowledge structures, such as ontologies, can provide valuable domain knowledge and user information. However, acquiring such knowledge and keeping it up to date is not a trivial task and user interests are particularly difficult to acquire and maintain. This paper investigates the synergy between a web-based research paper recommender system and an ontology containing information automatically extracted from departmental databases available on the web. The ontology is used to address the recommender systems cold-start problem. The recommender system addresses the ontology's interest-acquisition problem. An empirical evaluation of this approach is conducted and the performance of the integrated systems measured.

Artequakt: Generating Tailored Biographies from Automatically Annotated Fragments from the Web

Abstract: The Artequakt project seeks to automatically generate narrative biographies of artists from knowledge that has been extracted from the Web and maintained in a knowledge base. An overview of the system architecture is presented here and the three key components of that architecture are explained in detail, namely knowledge extraction, information management and biography construction. Conclusions are drawn from the initial experiences of the project and future progress is detailed.

CoAKTinG: Collaborative Advanced Knowledge Technologies in the Grid

Abstract: Grid infrastructures coupled with semantic web linkage and reasoning open up intriguing new possibilities for scientific collaboration. In this short paper, we outline the research agenda and collaboration technologies under development within the CoAKTinG project: Collaborative Advanced Knowledge Technologies in the Grid. CoAKTinG will provide tools to assist scientific collaboration by integrating intelligent meeting spaces, ontologically annotated media streams from online meetings, decision rationale and group memory capture, meeting facilitation, issue handling, planning and coordination support, constraint satisfaction, and instant messaging/presence. Their integration is illustrated through an extended use scenario.

ONTOCOPI: Methods and Tools for Identifying Communities of Practice

Abstract: The paper describes ONTOCOPI, a tool for identifying communities of practice (COPs) by analysing ontologies of the relevant working domain. COP identification is currently a resource-heavy process largely based on interviews. ONTOCOPI attempts to uncover informal COP relations by spotting patterns in the formal relations represented in ontologies, traversing the ontology from instance to instance via selected relations. Experiments to determine particular COPs from an academic ontology are described, showing how the alteration of threshold and temporal settings, and the weights applied to the ontology?s relations affect the composition of the identified COP.

Managing Reference: Ensuring Referential Integrity of Ontologies for the Semantic Web

Abstract: The diversity and distributed nature of the resources available in the semantic web poses significant challenges when these are used to help automatically build an ontology. One persistent and pervasive problem is that of the resolution or elimination of coreference that arises when more than one identifier is used to refer to the same resource. Tackling this problem is crucial for the referential integrity, and subsequently the quality of results, of any ontology-based knowledge service. We have built a coreference management service to be used alongside the population and maintenance of an ontology. An ontology based knowledge service that identifies communities of practice (CoPs) is also used to maintain the heuristics used in the coreference management system. This approach is currently being applied in a large scale experiment harvesting resources from various UK computer science departments with the aim of building a large, generic web-accessible ontology.

Identifying Communities of Practice: Analysing Ontologies as Networks to Support Community Recognition

Abstract: Communities of practice are seen as increasingly important for creating, sharing and applying organisational knowledge. Yet their informal nature makes them difficult to identify and manage. In this paper we set out ONTOCOPI, a system that applies ontology-based network analysis techniques to target the problem of identifying such communities.

Multiplicative synaptic normalization and a nonlinear Hebb rule underlie a neurotrophic model of competitive synaptic plasticity

Abstract: Synaptic normalization is used to enforce competitive dynamics in many models of developmental synaptic plasticity. In linear and semilinear Hebbian models, multiplicative synaptic normalization fails to segregate afferents whose activity patterns are positively correlated. To achieve this, the biologically problematic device of subtractive synaptic normalization must be used instead. Our own model of competition for neurotrophic support, which can segregate positively correlated afferents, was developed in part in an attempt to overcome these problems by removing the need for synaptic normalization altogether. However, we now show that the dynamics of our model decompose into two decoupled subspaces, with competitive dynamics being implemented in one of them through a nonlinear Hebb rule and multiplicative synaptic normalization. This normalization is "emergent" rather than imposed. We argue that these observations permit biologically plausible forms of synaptic normalization to be viewed as abstract and general descriptions of the underlying biology in certain scaleless models of synaptic plasticity.

Identifying Communities of Practice

Abstract: Communities of practice are seen as increasingly important for creating, sharing and applying organisational knowledge. Yet their informal nature makes them difficult to identify and manage. In this paper we set out ONTOCOPI, a system that applies ontology-based network analysis techniques to target the problem of identifying such communities.

Working smarter, not harder

Abstract: 2 1094-7167/02/$17.00 ©2002 IEEE IEEE INTELLIGENT SYSTEMS skip the first one because I could not find the time! I believed that my TM problems were due to a lack of moral fiber and could easily be sorted out with a bit more selfdiscipline. As I looked around at my fellow students, I wondered whether academics were just intrinsically chaotic, or hopeless workaholics, or both. But even the most work-addicted of us must, on occasion, wish they could get through that interminable email inbox, finish that paper, get around to writing that grant proposal.... By the end of the day I was pleasantly surprised—we were neither disorganized nor chaotic, we weren’t workaholics, and we weren’t entirely dysfunctional. We were simply dealing with information and task overload. Moreover, everyone appears to be in the same boat. Everyone is suffering the same deluge of information, requests, and tasks. The concerns and anxieties that I expressed were the same for the other professors, researchers, managers, administrators, and support staff. I’d be surprised if things were different in any 21st-century organizational setting. The issues are so pervasive and recurrent that, after a day of brainstorming, we produced responses that were almost identical to those in the course handouts. The handouts had been culled from countless other courses before us. The contexts, problems, and possible solutions were all there. This was reusable knowledge, and it set me thinking about the extent to which IT in general and intelligent systems in particular have contributed to the problem. It also led me to speculate about the prospects for IS helping in this area. Maxed out One of the most insidious TM problems, we learned, is perfectionism. We must “avoid letting the perfect be the enemy of the good.” Neither do we delegate or think about delegation sufficiently. We tell ourselves that explaining the task to someone else will take so long that doing it ourselves will be quicker. Another familiar TM problem is the information overload I mentioned earlier. There is now so much information that making a timely decision can be difficult. For example, the excellent How Much Information? project at the University of California, Berkeley, (www.sims. berkeley.edu/research/projects/how-much-info) calculated that in 2000, approximately 600 billion email messages were sent, although the researchers thought this figure was conservative. With an average size of 18,500 bytes per message and growing, the amount of material being pushed around in email alone was somewhere between 11,285 and 20,350 Tbytes. The Economist calculated that in 1999 the total world production of information in optical, paper, film, and magnetic media was 2,120,000 Tbytes. With all this information, a key TM challenge has got to be finding the right stuff. A recent study revealed that top designers in one world-renowned manufacturing company spent 40 percent of their time looking for information. But information overload and locating the information you need when you need it are not the only problems in trying to work smarter. We are constantly being interrupted in our tasks. We are expected to interleave or multitask a number of jobs during the day, and we seldom have time for much context switching between them. And then there is the old suspicion that we have simply too much to do. The list goes on, but it is interesting that many issues raised here are exactly those that drive knowledge management research. Recently, I attended a time management course run by our university. I have rather a jaundiced view of

Engineering knowledge for engineering grid applications

Abstract: Computing increasingly addresses collaboration; sharing; and interaction involving distributed resources. This has been fuelled in part by the emergence of Grid technologies and web services. Drawing on our expertise in the Geodise project. We argue that there is a growing requirement for knowledge engineering methods that provide a semantic foundation for such distributed computing. Such methods also support the sharing and coordinated use of knowledge itself. In this paper we introduce a service-oriented knowledge engineering approach that seeks to provide knowledge orientated support for distributed grid-based computing. This approach has been implemented in a generic integrated architecture. The application context is the process of design search and optimisation in engineering. It demonstrates how knowledge has been captured and modelled, as well as illustrating how ontologies have been developed and deployed. The knowledge acquired has been made available and accessible through a portal that invokes a number of basic services.

APECKS: using and evaluating a tool for ontology construction with internal and external KA support

Abstract: This paper describes Adaptive Presentation Environment for Collaboration Knowledge Structuring (APECKS), an experimental tool for collaborative ontology construction. APECKS takes a different line to most ontology servers, in that it is designed for use by domain experts, possibly in the absence of a knowledge engineer, and its aim is to foster and support debate about domain ontologies. To that end, it does not enforce ideals of consistency or correctness, and instead allows different conceptualizations of a domain to coexist. The system architecture and life cycle are introduced, and three extensive scenarios are outlined, showing how APECKS supports ontology construction, learning, ontology comparison and discussion. APECKS has also been used by several subjects during an evaluation experiment, and the results of this experiment are described. A particular factor about APECKS is that, as well as providing internal KA support, it is designed to interface with web-accessible KA tools, thereby allowing theoretically unlimited KA support for users. The prototype used WebGrid-II as external KA support, and the issues involved in integrating APECKS and WebGrid are discussed in detail.

Grid Enabled Optimisation and Design Search (Geodise)

Abstract: During the process of optimisation and design search, the modelling nad analysis of engineering problems are explited to yield improved designs The engineer explores various design parameters that he wishes to optimise and a measure of the quality of a particular design (the objective function) is computed using an appropriate model A number of algorithms may be used to yield more information about the behaviour of a model, and to minimise/maximise the objective function, and hence improve the quality of the design This process may include lengthy and repetitive calculations to obtain the value of the objective function with respect to the design variables

Foxtrot Recommender System: User profiling, Ontologies and the World Wide Web

Abstract: The Foxtrot recommender system recommends on-line research papers from a dynamic database. An ontology is used to represent user profiles, allowing ontological relationships to infer more information about a profile than can be observed directly from user behaviour and the use of a shared training set for multi-class classification. Profiles are also visualized in terms users understand, allowing direct profile feedback. A year long trial with over 300 people is currently underway to evaluate this system, while it provides a real world service to the staff and students of Southampton University.

Initiating Organizational Memories Using Ontology Network Analysis

Abstract: One of the important problems in organizational memories is their initial set-up. It is difficult to choose the right information to include in an organizational memory, and the right information is also a prerequisite for maximizing the uptake and relevance of the memory content. To tackle this problem, most developers adopt heavy-weight solutions and rely on a faithful continuous interaction with users to create and improve its content. In this paper, we explore the use of an automatic, light-weight solution, drawn from the underlying ingredients of an organizational memory: ontologies. We have developed an ontology-based network analysis method which we applied to tackle the problem of identifying communities of practice in an organization. We use ontology-based network analysis as a means to provide content automatically for the initial set-up of an organizational memory.

Dissociating ocular dominance column development and ocular dominance plasticity: a neurotrophic model.

Abstract: Recent experimental data indicate that both neurotrophic factors (NTFs) and intracortical inhibitory circuitry are implicated in the development and plasticity of ocular dominance columns. We extend a neurotrophic model of developmental synaptic plasticity, which previously failed to account correctly for the differences between monocular deprivation and binocular deprivation, and show that the inclusion of lateral cortical inhibition is indeed necessary in understanding the effects of visual deprivation in the model. In particular, we argue that monocular deprivation causes a differential shift in the balance between inhibition and excitation in cortical columns, down-regulating NTFs in deprived-eye columns and up-regulating NTFs in undeprived-eye columns; during binocular deprivation, however, no such shift occurs. We thus postulate that the response to visual deprivation is at the level of the cortical circuit, while the mechanisms of afferent segregation are at the molecular or cellular level. Such a dissociation is supported by recent experimental work challenging the assumption that columnar organisation develops in an activity-dependent, competitive fashion. Our extended model also questions recent attempts to distinguish between heterosynaptic and homosynaptic models of synaptic plasticity.

Initiating Organizational Memories using Ontology-based Network Analysis as a Bootstrapping Tool

Abstract: An important problem for many kinds of knowledge systems is their initial set-up. It is difficult to choose the right information to include in such systems, and the right information is also a prerequisite for maximizing the uptake and relevance. To tackle this problem, most developers adopt heavyweight solutions and rely on a faithful continuous interaction with users to create and improve content. In this paper, we explore the use of an automatic, lightweight ontology-based solution to the bootstrapping problem, in which domain-describing ontologies are analysed to uncover significant yet implicit relationships between instances. We illustrate the approach by using such an analysis to provide content automatically for the initial set-up of an organizational memory.

Knowledge Services for Distributed Service Integration

Abstract: Key Objectives: Introduce a service-oriented approach to providing knowledge support for distributed computing, develop a generic knowledge service architecture to realise such an approach, apply this approach design search and optimisation (Geodise) to enhance the design process and also for architecture validation. Motivation for the work: While computing increasingly addresses collaboration, sharing and interaction involve distributed services, there is a growing demand for knowledge services that provide underlying semantic support for such distributed services and also support the sharing and coordinated use of knowledge itself.

Capturing Knowledge of User Preferences: ontologies on recommender systems

Abstract: Tools for filtering the World Wide Web exist, but they are hampered by the difficulty of capturing user preferences in such a dynamic environment. We explore the acquisition of user profiles by unobtrusive monitoring of browsing behaviour and application of supervised machine-learning techniques coupled with an ontological representation to extract user preferences. A multi-class approach to paper classification is used, allowing the paper topic taxonomy to be utilised during profile construction. The Quickstep recommender system is presented and two empirical studies evaluate it in a real work setting, measuring the effectiveness of using a hierarchical topic ontology compared with an extendable flat list.

Grid services in action: grid enabled optimisation and design search

Abstract: We are developing a Grid Enabled Optimisation and Design Search system (GEODISE). It offers grid-based access to a state-of-the-art collection of optimisation and design search tools, industrial strength analysis codes, and distributed computing and data resources.

Who does the science in E-science?

Abstract: In my July/August 2001 editorial, I discussed how grid-based computing and the e-science it supports could powerfully influence the direction, development, and deployment of intelligent systems. E-science offers a promising vision of how computer and communication technology can support and enhance the scientific process. It does this by enabling scientists to more effectively generate, analyze, share, and discuss their insights, experiments, and results.

ONTOCOPI and Tools for Identifying Communities of Practice

Abstract: The paper describes ONTOCOPI, a tool for identifYing communities of practice (COPs) by analysing ontologies of the relevant working domain. COP identification is currently a resource-heavy process largely based on interviews. ONTOCOPI attempts to uncover informal COP relations by spotting patterns in the formal relations represented in ontologies, traversing the ontology from instance to instance via selected relations. Experiments to determine particular COPs from an academic ontology are described, showing how the alteration of threshold and temporal settings, and the weights applied to the ontology's relations affect the composition ofthe identified COP.