Abstract
Most of the work in machine learning assume that examples are generated at random according to some stationary probability distribution. In this work we study the problem of learning when the distribution that generate the examples changes over time. We present a method for detection of changes in the probability distribution of examples. The idea behind the drift detection method is to control the online error-rate of the algorithm. The training examples are presented in sequence. When a new training example is available, it is classified using the actual model. Statistical theory guarantees that while the distribution is stationary, the error will decrease. When the distribution changes, the error will increase. The method controls the trace of the online error of the algorithm. For the actual context we define a warning level, and a drift level. A new context is declared, if in a sequence of examples, the error increases reaching the warning level at example k(w), and the drift level at example k(d). This is an indication of a change in the distribution of the examples. The algorithm learns a new model using only the examples since k(w). The method was tested with a set of eight artificial datasets and a real world dataset. We used three learning algorithms: a perceptron, a neural network and a decision tree. The experimental results show a good performance detecting drift and with learning the new concept. We also observe that the method is independent of the learning algorithm.

Abstract
This paper presents the Simulator of Intelligent Transportation Systems (SITS). The SITS is based on a microscopic simulation approach to reproduce real traffic conditions in an urban or non-urban network and considers different types of vehicles, drivers and roads. A dynamical analysis of several traffic phenomena is then addressed. The results of using classical system theory tools point out that it is possible to study traffic systems, taking advantage on the knowledge gathered with automatic control algorithms. In this line of thought, it was also presented a new modelling formalism based on the embedding of statistics and Fourier transform.

Abstract
The paper presents the simulator of intelligent transportation systems (SITS). The SITS is based on a microscopic simulation approach to reproduce real traffic conditions in an urban or nonurban network and considers different types of vehicles, drivers, and roads. A dynamical analysis of several traffic phenomena is then addressed. The results of using classical system theory tools notes that it is possible to study traffic systems, taking advantage of the knowledge gathered with automatic control algorithms. In this line of thought, a new modeling formalism based on the embedding of statistics and Fourier transform is also presented.

Abstract
This paper presents an hybrid adaptive system for induction of forest of trees from data streams. The Ultra Fast Forest Tree system (UFFT) is an incremental algorithm, with constant time for processing each example, works online, and uses the Hoeffding bound to decide when to install a splitting test in a leaf leading to a decision node. Our system has been designed for continuous data. It uses analytical techniques to choose the splitting criteria, and the information gain to estimate the merit of each possible splitting-test. The number of examples required to evaluate the splitting criteria is sound, based on the Hoeffding bound. For multiclass problems,the algorithm builds a binary tree for each possible pair of classes, leading to a forest of trees. During the training phase the algorithm maintains a short term memory. Given a data stream, a fixed number of the most recent examples are maintained in a data-structure that supports constant time insertion and deletion. When a test is installed, a leaf is transformed into a decision node with two descendant leaves. The sufficient statistics of these leaves are initialized with the examples in the short term memory that will fall at these leaves. We study the behavior of UFFT in different problems. The experimental results shows that UFFT is competitive against a batch decision tree learner in large and medium datasets.

Abstract
Function allocation, as a process used in the construction of dependable complex systems, is a significant aspect of the design and implementation of interactive systems. It involves a documented and rational process for deciding what aspects of the system should be controlled by which human roles in the system and how the system should be automated to support these roles effectively. As computer systems have become more advanced, and the control of systems more complex, the notion of dynamic function allocation becomes increasingly desirable where in certain situations the automation may take over or give back function to the human user. In this paper we explore a further variant of dynamic function allocation that reflects typical work activity where the dynamic scheduling of activities takes place on the time dimension. The paper discusses this approach to dynamic function allocation called dynamic function scheduling and discusses the role that timed model checking may play in helping identify dependable dynamic function scheduling solutions.

Abstract
It is presented a promising approach to test interactive components, supporting the automatic generation of test cases from a specification. The relevance and difficulties (issues and challenges) associated with the testing of interactive components are first presented. It is shown that a formal specification with certain characteristics allows the automatic generation of test cases while solving some of the issues presented. The approach is illustrated with an example of automatic testing of the conformity between the implementation of a button, in the .Net framework, and a specification, written in the AsmL language, using the AsmL Tester tool.The conclusion discusses the characteristics of the tool and gives directions for future work.