In this thesis we present a system for detection of events in video. First a multiview approach to automatically detect and track heads and hands in a scene is described. Then, by making use of epipolar, spatial, trajectory, and appearance constraints, objects are labeled consistently across cameras (zooms). Finally, we demonstrate a new machine learning paradigm, TemporalBoost, that can recognize events in video. One aspect of any machine learning algorithm is in the feature set used. The approach taken here is to build a large set of activity features, though TemporalBoost itself is able to work with any feature set other boosting algorithms use. We also show how multiple levels of zoom can cooperate to solve problems related to activity recognition.

No description available.

No description available.

No description available.

Although nondestructive evaluation techniques have been implemented in many industry fields and proved to be useful, they are generally expensive, time consuming, and the results may not always be reliable. To overcome these drawbacks, structural health monitoring (SHM) systems has received significant attention in the past two decades. As structural systems are becoming more complicated and new materials are being developed, new methodologies, theories, and approaches in SHM have been developed for damage detection, diagnosis, and prognosis. Among the methods developed, the guided Lamb wave based SHM can be a promising technique for damage evaluation since it provides reliable damage information through signals propagating over large distance with little loss of amplitude. While this method is effective for damage assessment, the guided Lamb wave contains complicated mode characteristics, i.e. an infinite number of wave modes exist and these modes are generally dispersive. For this reason, a minimum number of wave modes and various signal processing algorithms are implemented to obtain better signal interpretations. Phased array beamsteering is an effective means for damage detection in guided Lamb wave SHM systems. Using this method, the wave energy can be focused at localized directions or areas by controlled excitation time delay of each array element. In this research, two types of transducers are utilized as phased array elements to compare beamsteering characteristics. Monolithic piezoceramic (PZT) transducers are investigated for beamsteering by assuming omnidirectional point sources for each actuator. MacroFiber Composite (MFC) transducers with anisotropic actuation are also studied, considering the wave main lobe width, main lobe magnitude, and side lobe levels. Analysis results demonstrate that the MFC phased arrays perform better than the PZT phased arrays for a range of beamsteering angles and have reduced main lobe width and side lobe levels. Experiments using the PZT and MFC phased arrays on an aluminum plate are also performed and compared to the analysis results. A time-frequency signal processing algorithm coupled with a machine learning method can form a robust damage diagnostic system. Four types of such algorithms, i.e. short time Fourier transform, Wigner-Ville distribution, wavelet transform, and matching pursuit, are investigated to select an appropriate algorithm for damage classification, and a spectrogram based on short time Fourier transform is adopted for its suitability. A machine learning algorithm called Adaboost is chosen due to its effectiveness and high accuracy performance. The classification is preformed using spectrograms and Adaboost for crack and corrosion damages. Artificial cracks and corrosions are created in Abaqus® to obtain the training samples consist of spectrograms. Several beam experiments in laboratory and additional simulations are also performed to get the testing samples for Adaboost. The analysis results show that not only correct damage classification is possible, but the confidence levels of each sample are acquired. / Ph. D.

No / Developing advanced and efficient malware detection systems is becoming significant in light of the growing threat landscape in cybersecurity. This work aims to tackle the enduring problem of identifying malware and protecting digital assets from cyber-attacks. Conventional methods frequently prove ineffective in adjusting to the ever-evolving field of harmful activity. As such, novel approaches that improve precision while simultaneously taking into account the ever-changing landscape of modern cybersecurity problems are needed. To address this problem this research focuses on the detection of malware in network traffic. This work proposes a machine-learning-based approach for malware detection, with particular attention to the Random Forest (RF), Support Vector Machine (SVM), and Adaboost algorithms. In this paper, the model’s performance was evaluated using an assessment matrix. Included the Accuracy (AC) for overall performance, Precision (PC) for positive predicted values, Recall Score (RS) for genuine positives, and the F1 Score (SC) for a balanced viewpoint. A performance comparison has been performed and the results reveal that the built model utilizing Adaboost has the best performance. The TPR for the three classifiers performs over 97% and the FPR performs < 4% for each of the classifiers. The created model in this paper has the potential to help organizations or experts anticipate and handle malware. The proposed model can be used to make forecasts and provide management solutions in the network’s everyday operational activities.

[pt] Boosting é uma técnica de aprendizado de máquina que combina diversos classificadores fracos com o objetivo de melhorar a acurácia geral. Em cada iteração, o algoritmo atualiza os pesos dos exemplos e constrói um classificador adicional. Um esquema simples de votação é utilizado para combinar os classificadores. O algoritmo mais famoso baseado em Boosting é o AdaBoost. Este algoritmo aumenta os pesos dos exemplos em que os classificadores anteriores cometeram erros. Assim, foca o classificador adicional nos exemplos mais difíceis. Inicialmente, uma distribuição uniforme de pesos é atribúda aos exemplos. Entretanto, não existe garantia que essa seja a melhor escolha para a distribuição inicial. Neste trabalho, apresentamos o Boosting at Start (BAS), uma nova abordagem de aprendizado de máquina baseada em Boosting. O BAS generaliza o AdaBoost permitindo a utilização de uma distribuição inicial arbitrária. Também apresentamos esquemas para determinação de tal distribuição. Além disso, mostramos como adaptar o BAS para esquemas de Aprendizado Semi-supervisionado. Adicionalmente, descrevemos a aplicação do BAS em diferentes problemas de classificação de dados e de texto, comparando o seu desempenho com o algoritmo AdaBoost original e alguns algoritmos do estado-da-arte para tais tarefas. Os resultados experimentais indicam que uma modelagem simples usando o algoritmo BAS gera classificadores eficazes. / [en] Boosting is a Machine Learning technique that combines several weak classifers with the goal of improving the overall accuracy. In each iteration, the algorithm updates the example weights and builds an additional classifer. A simple voting scheme is used to combine the classifers. The most famous Boosting-based algorithm is AdaBoost. This algorithm increases the weights of the examples that were misclassifed by the previous classifers. Thus, it focuses the additional classifer on the hardest examples. Initially, an uniform weight distribution is assigned to the examples. However, there is no guarantee that this is the best choice for the initial distribution. In this work, we present Boosting at Start (BAS), a new Machine Learning approach based on Boosting. BAS generalizes AdaBoost by allowing the use of an arbitrary initial distribution. We present schemes for the determination of such distribution. We also show how to adapt BAS to Semi-supervised learning schemes. Additionally, we describe the application of BAS in different problems of data and text classifcation, comparing its performance with the original AdaBoost algorithm and some state-of-the-art algorithms for such tasks. The experimental results indicate that a simple modelling using the BAS algorithm generates effective classifers.

Maskininlärning (ML) är idag ett mycket aktuellt, populärt och aktivt forskat område. Därav finns det idag en stor uppsjö av olika avancerade och moderna ML-algoritmer. Svårigheten är att bland dessa identifiera den mest optimala att applicera på ens tillämpningsområde. Algoritmer som bygger på Gradientförstärkning (eng. Gradient Boosting (GB)) har visat sig ha ett väldigt brett spektrum av appliceringsområden, flexibilitet, hög förutsägelseprestanda samt låga tränings- och förutsägelsetider. Huvudsyftet med denna studie är på klassificerings- och regressiondataset utvärdera och belysa prestandaskillnaderna av 5 moderna samt 2 äldre GB-algoritmer. Målet är att avgöra vilken av dessa moderna algoritmer som presterar i genomsnitt bäst utifrån på flera utvärderingsmått. Initialt utfördes en teoretisk förstudie inom det aktuella forskningsområdet. Algoritmerna XGBoost, LightGBM, CatBoost, AdaBoost, SnapBoost, KTBoost, GBDT implementerades på plattformen Google Colab. Där utvärderades dess respektive, tränings- och förutsägelsestid samt prestandamåtten, uppdelat i ROCAUC och Log Loss för klassificering samt R2 och RMSE för regression. Resultaten visade att det generellt var små skillnader mellan dom olika testade algoritmerna. Med undantag för AdaBoost som i allmänhet, med större marginal, hade den sämsta prestandan. Därmed gick det inte i denna jämförelse utse en klar vinnare. Däremot presterade SnapBoost väldigt bra på flera utvärderingsmått. Modellresultaten är generellt sätt väldigt begränsade och bundna till det applicerade datasetet vilket gör att det överlag är väldigt svårt att generalisera det till andra datauppsättningar. Detta speglar sig från resultaten med svårigheten att identifiera ett ML-ramverk som utmärker sig och presterar bra i alla scenarier. / Machine learning (ML) is today a very relevent, popular and actively researched area. As a result, today there exits a large numer of different advanced and modern ML algorithms. The difficulty is to identify among these the most optimal to apply to one’s area of application. Algorithms based on Gradient Boosting (GB) have been shown to have a very wide range of application areas, flexibility, high prediction performance and low training and prediction times. The main purpose of this study is on classification and regression datasets evaluate and illustrate the performance differences of 5 modern and 2 older GB algorithms. The goal is to determine which of these modern algorithms, on average, performs best on the basis of several evaluation metrics. Initially, a theoretical feasibility study was carried out in the current research area. The algorithms XGBoost, LightGBM, CatBoost, AdaBoost, SnapBoost, KTBoost, GBDT were implemented on the Google Colab platform. There, respective training and prediction time as well as the performance metrics were evaluated, divided into ROC-AUC and Log Loss for classification and R2 and RMSE for regression. The results showed that there were generally small differences between the different algorithms tested. With the exception of AdaBoost which in general, by a larger margin, had the worst performance. Thus, it was not possible in this comparison to nominate a clear winner. However, SnapBoost performed very well in several evaluation metrics. The model results are generally very limited and bound to the applied dataset, which makes it generally very difficult to generalize it to other data sets. This is reflected in the results with the difficulty of identifying an ML framework that excels and performs well in all scenarios.

The master's thesis deals with sense disambiguation of Czech words. Reader is informed about task's history and used algorithms are introduced. There are naive Bayes classifier, AdaBoost classifier, maximum entrophy method and decision trees described in this thesis. Used methods are clearly demonstrated. In the next parts of this thesis are used data also described.  Last part of the thesis describe reached results. There are some ideas to improve the system at the end of the thesis.

This thesis intends to evaluate the feasibility of supervised learning models for predicting house prices on the countryside of South Sweden. It is essential for mortgage lenders to have accurate housing valuation algorithms and the current model offered by Booli is not accurate enough when evaluating residence prices on the countryside. Different types of boosted decision trees were implemented to address this issue and their performances were compared to traditional machine learning methods. These different types of supervised learning models were implemented in order to find the best model with regards to relevant evaluation metrics such as root-mean-squared error (RMSE) and mean absolute percentage error (MAPE). The implemented models were ridge regression, lasso regression, random forest, AdaBoost, gradient boosting, CatBoost, XGBoost, and LightGBM. All these models were benchmarked against Booli's current housing valuation algorithms which are based on a k-NN model. The results from this thesis indicated that the LightGBM model is the optimal one as it had the best overall performance with respect to the chosen evaluation metrics. When comparing the LightGBM model to the benchmark, the performance was overall better, the LightGBM model had an RMSE score of 0.330 compared to 0.358 for the Booli model, indicating that there is a potential of using boosted decision trees to improve the predictive accuracy of residence prices on the countryside. / Denna uppsats ämnar utvärdera genomförbarheten hos olika övervakade inlärningsmodeller för att förutse huspriser på landsbygden i Södra Sverige. Det är viktigt för bostadslånsgivare att ha noggranna algoritmer när de värderar bostäder, den nuvarande modellen som Booli erbjuder har dålig precision när det gäller värderingar av bostäder på landsbygden. Olika typer av boostade beslutsträd implementerades för att ta itu med denna fråga och deras prestanda jämfördes med traditionella maskininlärningsmetoder. Dessa olika typer av övervakad inlärningsmodeller implementerades för att hitta den bästa modellen med avseende på relevanta prestationsmått som t.ex. root-mean-squared error (RMSE) och mean absolute percentage error (MAPE). De övervakade inlärningsmodellerna var ridge regression, lasso regression, random forest, AdaBoost, gradient boosting, CatBoost, XGBoost, and LightGBM. Samtliga algoritmers prestanda jämförs med Boolis nuvarande bostadsvärderingsalgoritm, som är baserade på en k-NN modell. Resultatet från denna uppsats visar att LightGBM modellen är den optimala modellen för att värdera husen på landsbygden eftersom den hade den bästa totala prestandan med avseende på de utvalda utvärderingsmetoderna. LightGBM modellen jämfördes med Booli modellen där prestandan av LightGBM modellen var i överlag bättre, där LightGBM modellen hade ett RMSE värde på 0.330 jämfört med Booli modellen som hade ett RMSE värde på 0.358. Vilket indikerar att det finns en potential att använda boostade beslutsträd för att förbättra noggrannheten i förutsägelserna av huspriser på landsbygden.