A major challenge for users of enterprise information is finding the information they want, presented in a way that makes sense to them. In this paper we tackle this problem by creating adaptive user profiles from implicit behaviour. In traditional approaches to information filtering, the user has to explicitly create his or her profile, and manually keep the profile up to date. Taking advantage of the popularity of collaborative tagging systems, we use the recorded tagging behaviour to construct an implicit, yet realistic and dynamic user profile. We present and evaluate algorithms for creating such profiles, characterizing their behaviour through statistical analysis. In addition, we present a visualisation tool which was used in a small scale user study to provide insight as to the effectiveness of our approach. Finally, we show how the profiles can be leveraged to enable personalised access to enterprise data sources.

In recent years the availability of GPS devices and the development in web technologies has produced a considerable growth in geographical applications available on the web. In particular the growing popularity of digital photography and photo sharing services has opened the way to a myriad of possible applications related to geotagged pictures. In this work we present an overview of the creation, sharing and use of geotagged pictures. We propose an approach to providing a new browsing experience of photo collections based on location and heading information metadata.

Artificial Immune Systems are engineering systems which have been inspired from the functioning of the biological immune system. We present an immune system model which incorporates two biologically motivated mechanisms to protect against autoimmune reactions, or false positives. The first, anergy, has been subject to the intense focus of immunologists as a possible key to autoimmune disease. The second is danger theory, which has attracted much interest as a possible alternative to traditional self-nonself selection models. We adopt a published immunological model, validate and extend it. Using the same calculations and assumptions as the original model, we integrate danger theory into the software. Without anergy, both models - the original and the danger model - produce similar results. When anergy is added, both models' performance improves. However, there seems to be some synergy between the mechanisms; anergy has a greater effect on the danger model than the original model. These findings should be of interest both to AIS practitioners and to the immunological community.

In recent years the growth in popularity of digital photography, together with the development of services and technologies to annotate and organize data on the Web, have extended the possibilities for managing and sharing large numbers of pictures. Our work explores the kinds of metadata that can be captured at the time a photo is taken, and ways to share these metadata in order to create a browsing experience of distributed photo collections based on their spatial information and relations. We present a prototype system in which an RDF description of pictures, including location and compass heading information, is used to discover geo-related pictures from other users. A browsing interface that allows users to explore pictures according to the spatial relationships discovered is proposed.

Semantic Blogging is the use of rich metadata to transform blogs from simple online diaries to full participants in an information sharing ecosystem. Originally the semantic blogging vision centred around informal decentralized knowledge management, but recent developments in social network analysis, microformats, semantic desktop applications and wikis have challenged, enriched and extended this vision. In this paper I review the history of semantic blogging, present a snapshot of where we are now, including two of my current experiments. BlogAccord is an exploration of music blogging, while the Snippet Manager is an information integration portal. I conclude this paper by offering a personal take on some promising future directions.

Gene libraries are a biological mechanism for generating combinatorial diversity in the immune system. However, they also bias the antibody creation process, so that they can be viewed as a way of guiding lifetime learning mechanisms. In this paper we examine the implications of this view, by examining coverage, avoidance of self, clustering and diversity. We show how gene libraries may improve both computational expense and performance, and present an analysis which suggests how they might do it. We suggest that gene libraries: provide combinatorial efficiency; improve coverage; reduce the cost of negative selection; and allow targeting of fixed antigen populations.

SWAN is a Semantic Web Application in Neuromedicine. SWAN is a project to develop an effective, integrated scientific knowledge infrastructure for the Alzheimer Disease (AD) research community, using the energy and self-organization of that community, enabled by Semantic Web technology. This infrastructure may later be deployed for research communities in other neuromedical disorders. SWAN incorporates the full biomedical research knowledge lifecycle in its ontological model, including support for personal data organization, hypothesis generation, experimentation, lab data organization, and digital pre-publication collaboration. Community, laboratory, and personal digital resources may all be organized and interconnected using SWANs common semantic framework

In recent years the availability of GPS devices and the development in web technologies has produced a considerable growth in geographical applications available on the web. In particular the growing popularity of digital photography and photo sharing services has opened the way to a myriad of possible applications related to geotagged pictures. In this work we present an overview of the creation, sharing and use of geotagged pictures. We propose an approach to providing a new browsing experience of photo collections based on location and heading information metadata.

Dendritic cells are antigen presenting cells that provide a vital link between the innate and adaptive immune system. Research into this family of cells has revealed that they perform the role of coordinating T-cell based immune responses, both reactive and for generating tolerance. We have derived an algorithm based on the functionality of these cells, and have used the signals and differentiation pathways to build a control mechanism for an artificial immune system. We present our algorithmic details in addition to some preliminary results, where the algorithm was applied for the purpose of anomaly detection. We hope that this algorithm will eventually become the key component within a large, distributed immune system, based on sound immunological concepts

Artificial Immune Systems (AIS) have been shown to be useful, practical and realisable approaches to real- world problems. Most AIS implementations are based around a canonical algorithm such as clonotypic learning, which we may think of as individual, lifetime learning. Yet a species also learns. Gene libraries are often thought of as a biological mechanism for generating combinatorial diversity of antibodies. However, they also bias the antibody creation process, so that they can be viewed as a way of guiding the lifetime learning mechanisms. Over time, the gene libraries in a species will evolve to an appropriate bias for the expected environment (based on species memory). Thus, gene libraries are a form of meta-learning which could be useful for AIS. Yet they are hardly ever used. In this paper we consider some of the possible benefits and implications of incorporating the evolution of gene libraries into AIS practice. We examine some of the issues that must be considered if the implementation is to be successful and beneficial

The semantic web has created various exciting opportunities to explore. Here we present a nature inspired solution to one such opportunity; that of semantic queries for information retrieval. We take our inspiration from the human immune system and develop an analogy between antibodies and queries. Successful antibodies are those that are activated by an infection. These antibodies are stimulated to clone, but imperfectly, giving rise to a multitude of similar antibodies that are better suited to tackle the infection. Analogously, queries producing relevant results can be cloned to give rise to various similar queries, each of which may be an improvement on the original query. The semantic web, being concept based, has a set of rules for creating expressive yet standardised queries with clear semantics guiding their modification. This paper discusses the implementation and evaluation of such an immune based information retrieval technique for the semantic web. Two query mutation operators; RandomMutationOperator and ConstrainedMutationOperator are proposed and compared in terms of their precision, recall and convergence. We have found the presented approach to be viable, and we discuss the potential for further improvements

The immune system is a complex biological system with a highly distributed, adaptive and self-organising nature. This paper presents an Artificial Immune System (AIS) that exploits some of these characteristics and is applied to the task of film recommendation by Collaborative Filtering (CF). Natural evolution and in particular the immune system have not been designed for classical optimisation. However, for this problem, we are not interested in finding a single optimum. Rather we intend to identify a sub-set of good matches on which recommendations can be based. It is our hypothesis that an AIS built on two central aspects of the biological immune system will be an ideal candidate to achieve this: Antigen-antibody interaction for matching and idiotypic antibody-antibody interaction for diversity. Computational results are presented in support of this conjecture and compared to those found by other CF techniques

Snippets are information nuggets that we would like to store, annotate and share. This is an example of what I call informal, decentralized knowledge management. Email is inadequate for this task; web logs (blogs) offer a decentralized though unstructured approach. I advocate the use of rich metadata to address this task; a technique I call Semantic Blogging

With the growing popularity of digital photography, amateur photographers face the prospect of having hundreds if not thousands of photos to label and organize. Current photo annotation tools allow users to organise their collections in sophisticated ways, and many people do this, but there is no way to share this organisational work over the Web. Combining RDF with a decentralized architecture similar to that used by blogs fosters the exchange of photographic metadata between heterogeneous tools and communities. In addition, the open nature of RDF means that photo sharers can contribute to, benefit from, and enrich resources that are exchanged within digital social networks such as blogs or FOAF.

In this paper, we describe the notion of a semantic information portal. This is a community information portal that exploits the semantic web standards to improve structure, extensibility, customisation and sustainability. We are in the process of developing a prototype directory of environmental organisations as a demonstration of the approach and outline the design challenges involved and the current status of the work

This paper is about semantic blogging, an application of the semantic web to blogging. The semantic web promises to make the web more useful by endowing metadata with machine processable semantics. Blogging is a lightweight web publishing paradigm which provides a very low barrier to entry, useful syndication and aggregation behaviour, a simple to understand structure and decentralised construction of a rich information network. Semantic blogging builds upon the success and clear network value of blogging by adding additional semantic structure to items shared over the blog channels

Over the last decade, a new idea challenging the classical self-non-self viewpoint has become popular amongst immunologists. It is called the Danger Theory. In this conceptual paper, we look at this theory from the perspective of Artificial Immune System practitioners

Immunologists are increasingly finding fault with traditional self-nonself thinking and a new <i>Danger Theory<i> (DT) is emerging. This new theory suggests that the immune system reacts to threats based on the correlation of various (danger) signals and it provides a method of groundingthe immune response, i.e. linking it directly to the attacker. Little is currently understood of the precise nature and correlation of these signals and the theory is a topic of hot debate. It is the aim of this research to investigate this correlation and to translate the DT into the realms of computer security, thereby creating AIS that are no longer limited by self-nonself discrimination

AIRS, a resource limited artificial immune classifier system, has performed well on elementary classification tasks. This paper proposes the use of this system for the more complex task of hierarchical, multi-class document classification

In this paper the construction and performance of a novel immune-based learning algorithm is explored. Through a process of cooperative co-evolution a classifier is generated which consists of a set of detectors whose local dynamics enable the system as a whole to group positive and negative examples of a concept. The immune-based learning algorithm is tested on a web-based document classification taskand found to outperform traditional classification paradigms

This paper sets out an approach which we call semantic blogging. We are building a demonstrator, set in the context of small group bibliography creation and management, which will illustrate the advantages of our approach

In this report we describe the lessons and experiences from developing a substantial semantic web application in the domain of community knowledge management

Over the last decade, a new idea challenging the classical self-non-self viewpoint has become popular amongst immunologists. It is called the Danger Theory. In this conceptual paper, we look at this theory from the perspective of Artificial Immune System practitioners. An overview of the Danger Theory is presented with particular emphasis on analogies in the Artificial Immune Systems world. A number of potential application areas are then used to provide a framing for a critical assessment of the concept, and its relevance for Artificial Immune Systems

It has previously been shown that a recommender based on immune system idiotypic principles can outperform one based on correlation alone. This paper reports the results of work in progress, where we undertake some investigations into the nature of this beneficial effect. The initial findings are that the immune system recommender tends to produce different neighbourhoods, and that the superior performance of this recommender is due partly to the different neighbourhoods, and partly to the way that the idiotypic effect is used to weight each neighbour’s recommendations.

This paper applies an artificial immune system (AIS) to the task of film recommendation by collaborative filtering (CF). It is our hypothesis that an AIS built on two central aspects of the biological immune system will be an ideal candidate to achieve this: Antigen - antibody interaction or matching and antibody - antibody interaction for diversity