Penny Noy

Multi-agent visualisation based on multivariate data

Interesting features of complex agent systems can be captured as multivariate data. There are a number of different approaches to visualizing such data. In this paper, we focus on methods which reduce the dimensions of the data through matrix transformations and then visualise the entities in the lower-dimensional space. We review an approach which describes agent similarities through distances, which are then visualised by multi-dimensional scaling techniques. We point out some shortcomings of this approach and examine an alternative, which applies principal component analysis and subsequent visualisation directly to the data. Our approach is implemented in the Space Explorer tool, which also allows interactive exploration. We identify four categories of data, which capture interaction, profiles, time series, and combinations of these three. Then we consider how to employ them for various agent types such as communicating, mobile, personal, interface, information and collaborating agents. Finally, we examine real-world telecoms data of 90,000 calls with Space Explorer.

Approaches to visualisation in bioinformatics: from dendrograms to Space Explorer

With the data explosion in biology visualisation techniques are of paramount importance for further progress. In this paper, we review traditional visualisation by clustering and dendrogram, which are prevailent in biology. We discuss its shortcomings and develop an alternative approach: Space Explorer 1 . In detail, we first present a framework to characterise the visualisation process. We identify two main data types and introduce structure comparison data and gene expression data as representatives, which serve as running examples throughout. Next, we review various distance measures and develop a design methodology for distances. We critically review the classical approach of clustering and visualisation through trees, in particular dendrograms, and pinpoint shortcomings of this technique. In order to tackle these shortcomings, we survey information visualisation techniques and we develop an alternative approach and system: Space Explorer, which maps the relationships of objects to distances and visualises these distances in a 3D, interactive space. We develop and evaluate three layout algorithms for the two data types and apply them to our case studies.

Creating Instruments for Ideation: Software Approaches to Geovisualization

Publisher Summary Ideation relates to the formation of ideas and concepts—the end goal of geovisualization. There are many tools and techniques for creating instruments for ideation—sophisticated hardware, advanced programming languages, graphics libraries, visual programming systems, and complex GUIs. In each, the developer or visualizer wishes to generate effective interactive graphic realizations of their data that are useful to them and/or their users. This chapter expands upon these ideas and considers the way each of these issues influences the uses and development of software instruments that support the exploratory process. Some examples of software approaches are also documented. Current technology has an important enabling and limiting impact upon the available range of instruments for ideation, which changes significantly over time. A major benefit of contemporary computer technology is the possibility to rapidly generate various graphical displays from data. This gives an opportunity to try alternative transient realizations of data, to discard those deemed ineffectual but when necessary reproduce them again, and to look at several displays simultaneously to provide multiple views of data. It is not only the increase of computer power that offers new opportunities to create more sophisticated instruments, but also the progress in software environments such as the development of programming tools that are high level and/or cross platform and the availability of libraries and reusable software components.

Defining like-minded agents with the aid of visualization

Profile carrying agents offer the opportunity of meeting like minds and increasing efficiency in many information search applications. Profiles can also increase the sophistication of relationships and interactions in multi-agent systems in general. Such profiles or feature lists may be of the agent owner or of the information sought. Two key issues are choice of similarity measure and privacy of profiles. In this paper we adapt techniques used in information visualization and classification to address these two problems. The basic idea is to map the high dimensional profiles into a low dimensional space. The mapping is one-way, so that privacy is achieved. To compare mapped profiles in the low dimensional space, we discuss the choice of a number of similarity measures.

Mobile Agents for Distributed Processing

In this statement, we sketch how to employ mobile agents for distributed computing and how such a solution compares to traditional approaches.

Creating Instruments for Ideation

Publisher Summary Ideation relates to the formation of ideas and concepts—the end goal of geovisualization. There are many tools and techniques for creating instruments for ideation—sophisticated hardware, advanced programming languages, graphics libraries, visual programming systems, and complex GUIs. In each, the developer or visualizer wishes to generate effective interactive graphic realizations of their data that are useful to them and/or their users. This chapter expands upon these ideas and considers the way each of these issues influences the uses and development of software instruments that support the exploratory process. Some examples of software approaches are also documented. Current technology has an important enabling and limiting impact upon the available range of instruments for ideation, which changes significantly over time. A major benefit of contemporary computer technology is the possibility to rapidly generate various graphical displays from data. This gives an opportunity to try alternative transient realizations of data, to discard those deemed ineffectual but when necessary reproduce them again, and to look at several displays simultaneously to provide multiple views of data. It is not only the increase of computer power that offers new opportunities to create more sophisticated instruments, but also the progress in software environments such as the development of programming tools that are high level and/or cross platform and the availability of libraries and reusable software components.

Advancing Profile Use in Agent Societies

Ubiquitous, persistent and pervasive computing systems and their modelling and engineering as agent societies suggest much potential for the application of personal and task profiles. Increasing quantities and types of profiling data are also becoming available. This paper discusses the use of profiles in this context and suggests a new form of profile employment to ease the use of profile data, by increasing convenience and privacy. Familiar dimension reduction techniques map the high dimensional profile to a position in a reduced dimension profile space. However, in this method, each profile owner calculates their position with respect to a number of reference vectors. This method avoids third party involvement and keeps the profile details private. Empirical studies indicate that calculating the whole set iteratively by this method compares favourably with respect to abstraction error when compared to direct calculation.