Agent software is a rapidly developing area of research. However, the overuse of the word “agent” has tended to mask the fact that, in reality, there is a truly heterogeneous body of research being carried out under this banner. This overview paper presents a typology of agents. Next, it places agents in context, defines them and then goes on, inter alia, to overview critically the rationales, hypotheses, goals, challenges and state-of-the-art demonstrators of the various agent types in our typology. Hence, it attempts to make explicit much of what is usually implicit in the agents literature. It also proceeds to overview some other general issues which pertain to all the types of agents in the typology. This paper largely reviews software agents, and it also contains some strong opinions that are not necessarily widely accepted by the agent community.

The basic nearest neighbour algorithm works by storing the training instances and classifying a new case by predicting that it has the same class as its nearest stored instance. To measure the distance between instances, some distance metric needs to be used. In situations when all attributes have numeric values, the conventional nearest neighbour method treats examples as points in feature spaces and uses Euclidean distance as the distance metric. In tasks with only nominal attributes, the simple “over-lap” metric is usually used. To handle classification tasks that have mixed types of attributes, the two different metrics are simply combined. Work by researchers in the machine learning field has shown that this approach performs poorly. This paper attempts to study a more recently developed distance metric and show that this metric is capable of measuring the importance of different attributes. With the use of discretisation for numeric-valued attributes, this method provides an integrated way in dealing with problem domains with mixtures of attribute types. Through detailed analyses, this paper tries to provide further insights into the understanding of nearest neighbour classification techniques and promote further use of this type of classification algorithm.

TFL, the Task Formal Language, has been developed for integrating the static and dynamic aspects of knowledge based systems. This paper focuses on the formal specification of dynamic behaviour. Although fundamental in knowledge based systems, strategic reasoning has been rather neglected until now by the existing formal specifications. Most languages were generally more focused on the domain and problem-solving knowledge specification than on the control. The formalisation presented here differs from previous ones in several aspects. First, a different representation of dynamic knowledge is proposed: TFL is based on Algebraic Data Types, as opposed to dynamic or temporal logic. Second, dynamic strategic reasoning is emphasised, whereas existing languages only offer to specify algorithmic control. Then, TFL does not only provide the specification of the problem-solving knowledge of the object system, but also of its strategic knowledge. Finally, the dynamic knowledge of the meta-system itself is also specified. Moreover, modularisation is another important feature of the presented language.

Time is one of the most relevant topics in AI. It plays a major role in several of AI research areas, ranging from logical foundations to applications of knowledge-based systems. Despite the ubiquity of time in AI, researchers tend to specialise and focus on time in particular contexts or applications, overlooking meaningful connections between different areas. In an attempt to promote crossfertilisation and reduce isolation, the Temporal Representation and Reasoning (TIME) workshop series was started in 1994. The third edition of the workshop was held on May 19–20 1996 in Key West, FL, with S. D. Goodwin and H. J. Hamilton as General Chairs, and L. Chittaro and A. Montanari as Program Chairs. A particular emphasis was given to the foundational aspects of temporal representation and reasoning through an investigation of the relationships between different approaches to temporal issues in AI, computer science and logic.

The Third International Workshop on Deontic Logic in Computer Science (ΔEON'96) took place in Sesimbra, Portugal, from 11–13 January 1996. It consisted of 12 refereed technical presentations and four invited talks. The invited speakers were Nuel Belnap (Pittsburgh University, USA), Andrew Jones (Oslo University, Norway), Krister Segerberg (Uppsala University, Sweden) and Marek Sergot (Imperial College, UK).

It is well known that a program—to accomplish any task—is welcomed by new users as long as it is easy to use, to understand, to learn. In the recent past, programs were written first and a nice adhoc interface was added later. The look and feel of this interface gradually became the convincing part of the whole system (program plus interface): in fact the way in which the screen presented the available commands, the possibility of on-line help, of undo operations, etc. were firstly greeted and then requested by the new users of complex programs.