Clustering aplicado à Bolsa de Valores de Lisboa

O objetivo desta dissertação foi estudar um conjunto de empresas cotadas na bolsa de valores de Lisboa, para identificar aquelas que têm um comportamento semelhante ao longo do tempo. Para isso utilizamos algoritmos de Clustering tais como K-Means, PAM, Modelos hierárquicos, Funny e C-Means tanto com a distância euclidiana como com a distância de Manhattan. Para selecionar o melhor número de clusters identificado por cada um dos algoritmos testados, recorremos a alguns índices de avaliação/validação de clusters como o Davies Bouldin e Calinski-Harabasz entre outros.

Detection and Identification of Abnormalities in Customer Consumptions in Power Distribution Systems

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

QK-Means: A clustering technique based on community detection and K-Means for deployment of cluster head nodes

Wireless Sensor Networks (WSN) are a special kind of ad-hoc networks that is usually deployed in a monitoring field in order to detect some physical phenomenon. Due to the low dependability of individual nodes, small radio coverage and large areas to be monitored, the organization of nodes in small clusters is generally used. Moreover, a large number of WSN nodes is usually deployed in the monitoring area to increase WSN dependability. Therefore, the best cluster head positioning is a desirable characteristic in a WSN. In this paper, we propose a hybrid clustering algorithm based on community detection in complex networks and traditional K-means clustering technique: the QK-Means algorithm. Simulation results show that QK-Means detect communities and sub-communities thus lost message rate is decreased and WSN coverage is increased. © 2012 IEEE.

Extracting Patterns from Educational Traces via Clustering and Associated Quality Metrics

Clustering algorithms, pattern mining techniques and associated quality metrics emerged as reliable methods for modeling learners’ performance, comprehension and interaction in given educational scenarios. The specificity of available data such as missing values, extreme values or outliers, creates a challenge to extract significant user models from an educational perspective. In this paper we introduce a pattern detection mechanism with-in our data analytics tool based on k-means clustering and on SSE, silhouette, Dunn index and Xi-Beni index quality metrics. Experiments performed on a dataset obtained from our online e-learning platform show that the extracted interaction patterns were representative in classifying learners. Furthermore, the performed monitoring activities created a strong basis for generating automatic feedback to learners in terms of their course participation, while relying on their previous performance. In addition, our analysis introduces automatic triggers that highlight learners who will potentially fail the course, enabling tutors to take timely actions.

Disaggregation of polygons of surficial geology and soil maps using spatial modelling and legacy data

Examples from the Murray-Darling basin in Australia are used to illustrate different methods of disaggregation of reconnaissance-scale maps. One approach for disaggregation revolves around the de-convolution of the soil-landscape paradigm elaborated during a soil survey. The descriptions of soil ma units and block diagrams in a soil survey report detail soil-landscape relationships or soil toposequences that can be used to disaggregate map units into component landscape elements. Toposequences can be visualised on a computer by combining soil maps with digital elevation data. Expert knowledge or statistics can be used to implement the disaggregation. Use of a restructuring element and k-means clustering are illustrated. Another approach to disaggregation uses training areas to develop rules to extrapolate detailed mapping into other, larger areas where detailed mapping is unavailable. A two-level decision tree example is presented. At one level, the decision tree method is used to capture mapping rules from the training area; at another level, it is used to define the domain over which those rules can be extrapolated. (C) 2001 Elsevier Science B.V. All rights reserved.

Spatial distribution of benthic microalgae on coral reefs determined by remote sensing

Understanding the ecological role of benthic microalgae, a highly productive component of coral reef ecosystems, requires information on their spatial distribution. The spatial extent of benthic microalgae on Heron Reef (southern Great Barrier Reef, Australia) was mapped using data from the Landsat 5 Thematic Mapper sensor. integrated with field measurements of sediment chlorophyll concentration and reflectance. Field-measured sediment chlorophyll concentrations. 2 ranging from 23-1.153 mg chl a m(2), were classified into low, medium, and high concentration classes (1-170, 171-290, and > 291 mg chl a m(-2)) using a K-means clustering algorithm. The mapping process assumed that areas in the Thematic Mapper image exhibiting similar reflectance levels in red and blue bands would correspond to areas of similar chlorophyll a levels. Regions of homogenous reflectance values corresponding to low, medium, and high chlorophyll levels were identified over the reef sediment zone by applying a standard image classification algorithm to the Thematic Mapper image. The resulting distribution map revealed large-scale ( > 1 km 2) patterns in chlorophyll a levels throughout the sediment zone of Heron Reef. Reef-wide estimates of chlorophyll a distribution indicate that benthic Microalgae may constitute up to 20% of the total benthic chlorophyll a at Heron Reef. and thus contribute significantly to total primary productivity on the reef.

Algorithms for time series clustering applied to biomedical signals

Thesis submitted in the fulfillment of the requirements for the Degree of Master in Biomedical Engineering

Algorithms for information extraction and signal annotation on long-term biosignals using clustering techniques

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

Origin and expansion of the allotetraploid Aegilops geniculata, a wild relative of wheat.

*This study reconstructs the phylogeography of Aegilops geniculata, an allotetraploid relative of wheat, to discuss the impact of past climate changes and recent human activities (e.g. the early expansion of agriculture) on the genetic diversity of ruderal plant species. *We combined chloroplast DNA (cpDNA) sequencing, analysed using statistical parsimony network, with nonhierarchical K-means clustering of amplified fragment length polymorphism (AFLP) genotyping, to unravel patterns of genetic structure across the native range of Ae. geniculata. The AFLP dataset was further explored by measurement of the regional genetic diversity and the detection of isolation by distance patterns. *Both cpDNA and AFLP suggest an eastern Mediterranean origin of Ae. geniculata. Two lineages have spread independently over northern and southern Mediterranean areas. Northern populations show low genetic diversity but strong phylogeographical structure among the main peninsulas, indicating a major influence of glacial cycles. By contrast, low genetic structuring and a high genetic diversity are detected in southern Mediterranean populations. Finally, we highlight human-mediated dispersal resulting in substantial introgression between resident and migrant populations. *We have shown that the evolutionary trajectories of ruderal plants can be similar to those of wild species, but are interfered by human activities, promoting range expansions through increased long-distance dispersal and the creation of suitable habitats.

A Topological Approach to Fuzzy Sets

This thesis presents a topological approach to studying fuzzy setsby means of modifier operators. Modifier operators are mathematical models, e.g., for hedges, and we present briefly different approaches to studying modifier operators. We are interested in compositional modifier operators, modifiers for short, and these modifiers depend on binary relations. We show that if a modifier depends on a reflexive and transitive binary relation on U, then there exists a unique topology on U such that this modifier is the closure operator in that topology. Also, if U is finite then there exists a lattice isomorphism between the class of all reflexive and transitive relations and the class of all topologies on U. We define topological similarity relation "≈" between L-fuzzy sets in an universe U, and show that the class LU/ ≈ is isomorphic with the class of all topologies on U, if U is finite and L is suitable. We consider finite bitopological spaces as approximation spaces, and we show that lower and upper approximations can be computed by means of α-level sets also in the case of equivalence relations. This means that approximations in the sense of Rough Set Theory can be computed by means of α-level sets. Finally, we present and application to data analysis: we study an approach to detecting dependencies of attributes in data base-like systems, called information systems.

Extra Forces induced by wide-pulse, high-frequency electrical stimulation: Occurrence, magnitude, variability and underlying mechanisms.

OBJECTIVE: In contrast to conventional (CONV) neuromuscular electrical stimulation (NMES), the use of "wide-pulse, high-frequencies" (WPHF) can generate higher forces than expected by the direct activation of motor axons alone. We aimed at investigating the occurrence, magnitude, variability and underlying neuromuscular mechanisms of these "Extra Forces" (EF). METHODS: Electrically-evoked isometric plantar flexion force was recorded in 42 healthy subjects. Additionally, twitch potentiation, H-reflex and M-wave responses were assessed in 13 participants. CONV (25Hz, 0.05ms) and WPHF (100Hz, 1ms) NMES consisted of five stimulation trains (20s on-90s off). RESULTS: K-means clustering analysis disclosed a responder rate of almost 60%. Within this group of responders, force significantly increased from 4% to 16% of the maximal voluntary contraction force and H-reflexes were depressed after WPHF NMES. In contrast, non-responders showed neither EF nor H-reflex depression. Twitch potentiation and resting EMG data were similar between groups. Interestingly, a large inter- and intrasubject variability of EF was observed. CONCLUSION: The responder percentage was overestimated in previous studies. SIGNIFICANCE: This study proposes a novel methodological framework for unraveling the neurophysiological mechanisms involved in EF and provides further evidence for a central contribution to EF in responders.

Responders to Wide-Pulse, High-Frequency Neuromuscular Electrical Stimulation Show Reduced Metabolic Demand: A 31P-MRS Study in Humans.

Conventional (CONV) neuromuscular electrical stimulation (NMES) (i.e., short pulse duration, low frequencies) induces a higher energetic response as compared to voluntary contractions (VOL). In contrast, wide-pulse, high-frequency (WPHF) NMES might elicit-at least in some subjects (i.e., responders)-a different motor unit recruitment compared to CONV that resembles the physiological muscle activation pattern of VOL. We therefore hypothesized that for these responder subjects, the metabolic demand of WPHF would be lower than CONV and comparable to VOL. 18 healthy subjects performed isometric plantar flexions at 10% of their maximal voluntary contraction force for CONV (25 Hz, 0.05 ms), WPHF (100 Hz, 1 ms) and VOL protocols. For each protocol, force time integral (FTI) was quantified and subjects were classified as responders and non-responders to WPHF based on k-means clustering analysis. Furthermore, a fatigue index based on FTI loss at the end of each protocol compared with the beginning of the protocol was calculated. Phosphocreatine depletion (ΔPCr) was assessed using 31P magnetic resonance spectroscopy. Responders developed four times higher FTI's during WPHF (99 ± 37 ×103 N.s) than non-responders (26 ± 12 ×103 N.s). For both responders and non-responders, CONV was metabolically more demanding than VOL when ΔPCr was expressed relative to the FTI. Only for the responder group, the ∆PCr/FTI ratio of WPHF (0.74 ± 0.19 M/N.s) was significantly lower compared to CONV (1.48 ± 0.46 M/N.s) but similar to VOL (0.65 ± 0.21 M/N.s). Moreover, the fatigue index was not different between WPHF (-16%) and CONV (-25%) for the responders. WPHF could therefore be considered as the less demanding NMES modality-at least in this subgroup of subjects-by possibly exhibiting a muscle activation pattern similar to VOL contractions.

Automatic Image Annotation Using SURF Descriptors

In recent years there is an apparent shift in research from content based image retrieval (CBIR) to automatic image annotation in order to bridge the gap between low level features and high level semantics of images. Automatic Image Annotation (AIA) techniques facilitate extraction of high level semantic concepts from images by machine learning techniques. Many AIA techniques use feature analysis as the first step to identify the objects in the image. However, the high dimensional image features make the performance of the system worse. This paper describes and evaluates an automatic image annotation framework which uses SURF descriptors to select right number of features and right features for annotation. The proposed framework uses a hybrid approach in which k-means clustering is used in the training phase and fuzzy K-NN classification in the annotation phase. The performance of the system is evaluated using standard metrics.

Comparing Support Vector Machines with Gaussian Kernels to Radial Basis Function Classifiers

The Support Vector (SV) machine is a novel type of learning machine, based on statistical learning theory, which contains polynomial classifiers, neural networks, and radial basis function (RBF) networks as special cases. In the RBF case, the SV algorithm automatically determines centers, weights and threshold such as to minimize an upper bound on the expected test error. The present study is devoted to an experimental comparison of these machines with a classical approach, where the centers are determined by $k$--means clustering and the weights are found using error backpropagation. We consider three machines, namely a classical RBF machine, an SV machine with Gaussian kernel, and a hybrid system with the centers determined by the SV method and the weights trained by error backpropagation. Our results show that on the US postal service database of handwritten digits, the SV machine achieves the highest test accuracy, followed by the hybrid approach. The SV approach is thus not only theoretically well--founded, but also superior in a practical application.