I would like to use this perspective to advocate further research in data mining toward methods that allow and formalize greater input from the modeler. By ‘modeler’, I mean the domain expert, scientist, business analyst, or researcher who has questions to answer using data. Soliciting input from such users is atypical of most data mining algorithms (e.g., see the introductory discussion in ref...

Software artifacts at different levels of abstraction are closely inter-related. Developers require support for managing these inter-relationships as artifacts evolve during development. We describe a conceptual architecture and prototype for supporting traceability and inconsistency management between software requirements descriptions, UML-style use case models and black-box test plans. Key i...

Explanations of modeling errors in ontologies are of crucial importance both when creating and maintaining the ontology. This work presents two novel incremental methods for error explanations in semantic web ontologies and shows that they have significantly better performance than the state of the art black-box techniques. Both promising techniques together with our implementation of a tableau...

Black-box optimization algorithms cannot use the specific parameters of the problem instance, i.e., of the fitness function f . Their run time is measured as the number of f -evaluations. This implies that the usual algorithmic complexity of a problem cannot be used in the black-box scenario. Therefore, a new framework for the valuation of algorithms for black-box optimization is presented allo...

Operating systems which implement a dynamic priority mechanism are very common. Nevertheless, it is very di cult to develop an accurate analytical model to evaluate their performance, mainly due to the different forms of dependency between the various constituent parts. In this paper, we introduce a black box modeling approach which allows us to identify and decompose the di erent functions of ...

In this paper, we present a mechanism able to predict the performance a given workload will achieve when running on a given storage device. This mechanism is composed by two modules. The first one is able to learn the behavior of a workload in order to be able to reproduce its behavior later on, without a new execution, even when the storage drives or data placement are modified. The second mod...

We welcome the Subcommittee’s review of “Air Pollution Modelling in Hong Kong”. Hong Kong’s air quality is a topic of great public concern. The lack of progress in improving air quality has led many to question the effectiveness of our air quality management system. Most people understand that air quality modelling is important for our air quality management system because of its role in air qu...

As network researchers who have spent many long days in the past few years poring over Gigabytes of networkrelated measurements to try to get some basic understanding of network traffic, we are very pleased to see this article and to be able to comment on it. Historically, the field of networking and communications research has suffered from a severe and constant shortage of traffic measurement...

The dynamic behaviour can then be derived from the steady state using a first order model. The steady state performances can accurately be assessed using black box modelling such as neural networks. Nowadays, neural networks are widely used to model nonlinear relationship between the data. However, even a black-box model needs few parameters to achieve its goal; the performance of its predictio...