Abstract
Most of supervised learning algorithms assume the stability of the target concept over time. Nevertheless in many real-user modeling systems, where the data is collected over an extended period of time, the learning task can be complicated by changes in the distribution underlying the data. This problem is known in machine learning as concept drift. The main idea behind Statistical Quality Control is to monitor the stability of one or more quality characteristics in a production process which generally shows some variation over time. In this paper we present a method for handling concept drift based on Shewhart P-Charts in an on-line framework for supervised learning. We explore the use of two alternatives P-charts, which differ only by the way they estimate the target value to set the center line. Experiments with simulated concept drift scenarios in the context of a user modeling prediction task compare the proposed method with other adaptive approaches. The results show that, both P-Charts consistently recognize concept changes, and that the learner can adapt quickly to these changes to maintain its performance level.

Abstract
In previous work (Brito and De Carvalho (1999)) we have considered the presence of dependence rules between variables in the framework of a symbolic clustering method. In another paper Brito (1998) has addressed the problem of clustering probabilistic data. The aim of this paper is to bring together the two issues, that is, to take into account dependence rules on probabilistic data. This is accomplished by introducing new generality measures with an appropriate generalization operator. This approach allows for the extension of a symbolic clustering. method to constrained probabilistic data.

Abstract
In this paper, the Unit Commitment (UC) problem is presented and solved, following an innovative approach based on a metaheuristic procedure. The problem consists on deciding which electric generators must be committed, over a given planning horizon, and on defining the production levels that are required for each generator, so that load and spinning reserve requirements are verified, at minimum production costs. Due to its complexity, exact methods proved to be inefficient when real size problems were considered. Therefore, heuristic methods have for long been developed and, in recent years, metaheuristics have also been applied with some success to the problem. Methods like Simulated Annealing, Tabu Search and Evolutionary Programming can be found in several papers, presenting results that are sufficiently interesting to justify further research in the area. In this paper, a resolution framework based on GRASP - Greedy Randomized Adaptive Search Procedure - is presented. To obtain a general optimisation tool, capable of solving different problem variants and of including several objectives, the operations involved in the optimisation process do not consider any particular characteristics of the classical UC problem. Even so, when applied to instances with very particular structures, the computational results show the potential of this approach.

Abstract
We present Adaptive Bayes, an adaptive incremental version of Naive Bayes, to model a prediction task based on learning styles in the context of an Adaptive Hypermedia Educational System. Since the student's preferences can change over time, this task is related to a problem known as concept drift in the machine learning community. For this class of problems an adaptive predictive model, able to adapt quickly to the user's changes, is desirable. The results from conducted experiments show that Adaptive Bayes seems to be a fine and simple choice for this kind of prediction task in user modeling.

Abstract
Three networks based on fused biconical wavelength division multiplexers (WDMs) and fiber Bragg gratings (FBGs) are theoretically and experimentally demonstrated for photonic-intensity-sensor multiplexing. The aim of replacing standard couplers for WDMs in the power division process is to reduce power losses and improve the robustness of the systems to FBG wavelength shifts. The different network topologies are analyzed both in terms of power budget and crosstalk noise, considering the multiplexing of two fiber-taper displacement sensors. The configuration with one detector for each sensor and the corresponding FBG at the detection end is proven to be the best topology in terms of crosstalk, doubling the peak-isolation value of the WDMs employed and yielding a 37.4-dB optical signal-to-noise ratio for a two-sensor network. Regarding power losses, the optimum configuration locates the FBGs at the sensor heads, thus improving power budget and avoiding additional couplers at the detectors. Both topologies are expanded to multiplex four sensors, being crosstalk identified as the critical factor in these networks. With this limiting parameter, the first configuration has been determined as the most suitable for multiplexing a high number of sensors.

Abstract
A micromachined low finesse Fabry-Perot interferometer for measuring dc and ac electrical current is presented. Interrogation of the microcavity is achieved by a dual-wavelength fiber Bragg grating technique working in quadrature. A linear relation between the dc electrical current and the optical phase defined by the microcavity was detected. Large enhancement of the sensitivity of the microcavities is presented with the use of a planar coil instead of a power line. The sensitivity of the sensor with the planar coil configuration is 7.9 rad/A and resolution of similar to0.18 mA/rootHz is achieved when the distance between the planar coil and the transducer head is 2 mm. The response of the sensor for ac measurements is 0.14 V/A with a resolution of 6 mA/rootHz when the distance between the power line and the transducer head is 5.5 cm.