This thesis is focused on object tracking by a decentralized sensor network using fusion center-based and consensus-based distributed particle filters. The model includes clutter as well as missed detections of the object. The approach uses sparsity of global likelihood function, which, by means of appropriate sparse approximation and the suitable dictionaty selection can significantly reduce communication requirements in the decentralized sensor network. The master's thesis contains a design of exchange methods of sparse data in the sensor network and a comparison of the proposed methods in terms of accuracy and energy requirements.

This diploma thesis deals with the detection of the current floor during elevator ride. This functionality is necessary for robot to move in multi-floor building. For this task, a fusion of accelerometric data during the ride of the elevator and image data obtained from the information display inside the elevator cabin is used. The research describes the already implemented solutions, data fusion methods and image classification options. Based on this part, suitable approaches for solving the problem were proposed. First, datasets from different types of elevator cabins were obtained. An algorithm for working with data from the accelerometric sensor was developed. A convolutional neural network, which was used to classify image data from displays, was selected and trained. Subsequently, the data fusion method was implemented. The individual parts were tested and evaluated. Based on their evaluation, integration into one functional system was performed. System was successfully verified and tested. Result of detection during the ride in different elevators was 97%.

The ability to recognize handshapes in signing video is essential in algorithms for sign recognition and retrieval. Handshape recognition from isolated images is, however, an insufficiently constrained problem. Many handshapes share similar 3D configurations and are indistinguishable for some hand orientations in 2D image projections. Additionally, significant differences in handshape appearance are induced by the articulated structure of the hand and variants produced by different signers. Linguistic rules involved in the production of signs impose strong constraints on the articulations of the hands, yet, little attention has been paid towards exploiting these constraints in previous works on sign recognition. Among the different classes of signs in any signed language, lexical signs constitute the prevalent class. Morphemes (or, meaningful units) for signs in this class involve a combination of particular handshapes, palm orientations, locations for articulation, and movement type. These are thus analyzed by many sign linguists as analogues of phonemes in spoken languages. Phonological constraints govern the ways in which phonemes combine in American Sign Language (ASL), as in other signed and spoken languages; utilizing these constraints for handshape recognition in ASL is the focus of the proposed thesis. Handshapes in monomorphemic lexical signs are specified at the start and end of the sign. The handshape transition within a sign are constrained to involve either closing or opening of the hand (i.e., constrained to exclusively use either folding or unfolding of the palm and one or more fingers). Furthermore, akin to allophonic variations in spoken languages, both inter- and intra- signer variations in the production of specific handshapes are observed. We propose a Bayesian network formulation to exploit handshape co-occurrence constraints also utilizing information about allophonic variations to aid in handshape recognition. We propose a fast non-rigid image alignment method to gain improved robustness to handshape appearance variations during computation of observation likelihoods in the Bayesian network. We evaluate our handshape recognition approach on a large dataset of monomorphemic lexical signs. We demonstrate that leveraging linguistic constraints on handshapes results in improved handshape recognition accuracy. As part of the overall project, we are collecting and preparing for dissemination a large corpus (three thousand signs from three native signers) of ASL video annotated with linguistic information such as glosses, morphological properties and variations, and start/end handshapes associated with each ASL sign.

Radar map generation using binary Bayes filter or what is commonly known as Inverse Sensor Model; which translates the sensor measurements into grid cells occupancy estimation, is a classical problem in different fields. In this work, the focus will be on development of Inverse Sensor Model for parking space using 77 GHz FMCW (Frequency Modulated Continuous Wave) automotive radar, that can handle different environment geometrical complexity in a parking space. There are two main types of Inverse Sensor Models, where each has its own assumption about the sensor noise. One that is fixed and is similar to a lookup table, and constructed based on combination of sensor-specific characteristics, experimental data and empirically-determined parameters. The other one is learned by using ground truth labeling of the grid map cell, to capture the desired Inverse Sensor Model. In this work a new Inverse Sensor Model is proposed, that make use of the computational advantage of using fixed Inverse Sensor Model and capturing desired occupancy estimation based on ground truth labeling. A derivation of the occupancy grid mapping problem using binary Bayes filtering would be performed from the well known SLAM (Simultaneous Localization and Mapping) problem, followed by presenting the Adaptive Inverse Sensor Model, that uses fixed occupancy estimation but with adaptive occupancy shape estimation based on statistical analysis of the radar measurements distribution across the acquisition environment. A prestudy of the noise nature of the radar used in this work is performed, to have a common Inverse Sensor Model as a benchmark. Then the drawbacks of such Inverse Sensor Model would be addressed as sub steps of Adaptive Inverse Sensor Model, to be able to haven an optimal grid map occupancy estimator. Finally a comparison between the generated maps using the benchmark and the adaptive Inverse Sensor Model will take place, to show that under the fulfillment of the assumptions of the Adaptive Inverse Sensor Model, the Adaptive Inverse Sensor Model can offer a better visual appealing map to that of the benchmark.

Text classification is a fundamental part of natural language processing. In this thesis, methods for text classification are used in an attempt to predict the political party affiliation of members of parliament (MPs). The objective is to evaluate the performance of Support Vector Machines (SVM), naive Bayes, and a fine-tuned Bidirectional Encoder Representations from Transformers (BERT) model in predicting MPs' political party affiliation based on speeches given in the Chamber of the Swedish Parliament. This study shows that BERT outperforms SVM and naive Bayes in correctly classifying MPs, and SVM makes better predictions than naive Bayes and performs reasonably well compared to BERT. The results show that all models correctly predict MPs representing the Sweden Democrats to the highest degree. Both BERT and SVM roughly classify every other speech correctly, which implies much better than making random predictions. These results indicate the potential use of methods for automatically classifying political speeches.

This project, conducted at Ericsson AB, investigates the feasibility of implementing machine learning techniques in order to classify dump files for more effi cient trouble report routing. The project focuses on supervised machine learning methods and in particular Bayesian statistics. It shows that a program utilizing Bayesian methods can achieve well above random prediction accuracy. It is therefore concluded that machine learning methods may indeed become a viable alternative to human classification of trouble reports in the near future. / Detta examensarbete, utfört på Ericsson AB, ämnar att undersöka huruvida maskininlärningstekniker kan användas för att klassificera dumpfiler för mer effektiv problemidentifiering. Projektet fokuserar på övervakad inlärning och då speciellt Bayesiansk klassificering. Arbetet visar att ett program som utnyttjar Bayesiansk klassificering kan uppnå en noggrannhet väl över slumpen. Arbetet indikerar att maskininlärningstekniker mycket väl kan komma att bli användbara alternativ till mänsklig klassificering av dumpfiler i en nära framtid.

Research philosophies and paradigms (RPPs) reveal researchers’ assumptions and provide a systematic way in which research can be carried out effectively and appropriately. Different studies highlight cognitive and comprehension challenges of RPPs concepts at the postgraduate level. This study develops a natural language processing (NLP) supervised classification application that guides students in identifying RPPs applicable to their study. By using algorithms rooted in a quantitative research approach, this study builds a corpus represented using the Bag of Words model to train the naïve Bayes, Logistic Regression, and Support Vector Machine algorithms. Computer experiments conducted to evaluate the performance of the algorithms reveal that the Naïve Bayes algorithm presents the highest accuracy and precision levels. In practice, user testing results show the varying impact of knowledge, performance, and effort expectancy. The findings contribute to the minimization of issues postgraduates encounter in identifying research philosophies and the underlying paradigms for their studies. / Science and Technology Education / MTech. (Information Technology)

The quality of a wine is important for the consumers as well as the wine industry. The traditional (expert) way of measuring wine quality is time-consuming. Nowadays, machine learning models are important tools to replace human tasks. In this case, there are several features to predict the wine quality but the entire features will not be relevant for better prediction. So, our thesis work is focusing on what wine features are important to get the promising result. For the purposeof classification model and evaluation of the relevant features, we used three algorithms namely support vector machine (SVM), naïve Bayes (NB), and artificial neural network (ANN). In this study, we used two wine quality datasets red wine and white wine. To evaluate the feature importance we used the Pearson coefficient correlation and performance measurement matrices such as accuracy, recall, precision, and f1 score for comparison of the machine learning algorithm. A grid search algorithm was applied to improve the model accuracy. Finally, we achieved the artificial neural network (ANN) algorithm has better prediction results than the Support Vector Machine (SVM) algorithm and the Naïve Bayes (NB) algorithm for both red wine and white wine datasets.

Cleavage and polyadenylation of a precursor mRNA is important for transcription termination, mRNA stability, and regulation of gene expression. This process is directed by a multitude of protein factors and cis elements in the pre-mRNA sequence surrounding the cleavage and polyadenylation site. Importantly, the location of the cleavage and polyadenylation site helps define the 3’ untranslated region of a transcript, which is important for regulation by microRNAs and RNA binding proteins. Additionally, these sites have generally been poorly annotated. To identify 3’ ends, many techniques utilize an oligo-dT primer to construct deep sequencing libraries. However, this approach can lead to identification of artifactual polyadenylation sites due to internal priming in homopolymeric stretches of adenines. Previously, simple heuristic filters relying on the number of adenines in the genomic sequence downstream of a putative polyadenylation site have been used to remove these sites of internal priming. However, these simple filters may not remove all sites of internal priming and may also exclude true polyadenylation sites. Therefore, I developed a naïve Bayes classifier to identify putative sites from oligo-dT primed 3’ end deep sequencing as true or false/internally primed. Notably, this algorithm uses a combination of sequence elements to distinguish between true and false sites. Finally, the resulting algorithm is highly accurate in multiple model systems and facilitates identification of novel polyadenylation sites.

Den stora mängden tillgänglig data på internet kan användas för att förbättra förutsägelser genom maskininlärning. Problemet är att sådan data ofta är i ett obehandlat format och kräver att någon manuellt bestämmer etiketter på den insamlade datan innan den kan användas av algoritmen. Semi-supervised learning (SSL) är en teknik där algoritmen använder ett fåtal förbehandlade exempel och därefter automatiskt bestämmer etiketter för resterande data. Ett tillvägagångssätt inom SSL är att representera datan i en graf, vilket kallas för graf-baserad semi-supervised learning (GSSL), och sedan hitta likheter mellan noderna i grafen för att automatiskt bestämma etiketter.Vårt mål i denna uppsatsen är att förenkla de avancerade processerna och stegen för att implementera en GSSL-algoritm. Vi kommer att gå igen grundläggande steg som hur utvecklingsmiljön ska installeras men även mer avancerade steg som data pre-processering och feature extraction. Feature extraction metoderna som uppsatsen använder sig av är bag-of-words (BOW) och term frequency-inverse document frequency (TF-IDF). Slutgiltligen presenterar vi klassificering av dokument med Label Propagation (LP) och Multinomial Naive Bayes (MNB) samt en detaljerad beskrivning över hur GSSL fungerar.Vi presenterar även prestanda för klassificering-algoritmerna genom att klassificera 20 Newsgroup datasetet med LP och MNB. Resultaten dokumenteras genom två olika utvärderingspoäng vilka är F1-score och accuracy. Vi gör även en jämförelse mellan MNB och LP med två olika typer av kärnor, KNN och RBF, på olika mängder av förbehandlade träningsdokument. Resultaten ifrån klassificering-algoritmerna visar att MNB är bättre på att klassificera datasetet än LP. / The large amount of available data on the web can be used to improve the predictions made by machine learning algorithms. The problem is that such data is often in a raw format and needs to be manually labeled by a human before it can be used by a machine learning algorithm. Semi-supervised learning (SSL) is a technique where the algorithm uses a few prepared samples to automatically prepare the rest of the data. One approach to SSL is to represent the data in a graph, also called graph-based semi-supervised learning (GSSL), and find similarities between the nodes for automatic labeling.Our goal in this thesis is to simplify the advanced processes and steps to implement a GSSL-algorithm. We will cover basic tasks such as setup of the developing environment and more advanced steps such as data preprocessing and feature extraction. The feature extraction techniques covered are bag-of-words (BOW) and term frequency-inverse document frequency (TF-IDF). Lastly, we present how to classify documents using Label Propagation (LP) and Multinomial Naive Bayes (MNB) with a detailed explanation of the inner workings of GSSL. We showcased the classification performance by classifying documents from the 20 Newsgroup dataset using LP and MNB. The results are documented using two different evaluation scores called F1-score and accuracy. A comparison between MNB and the LP-algorithm using two different types of kernels, KNN and RBF, was made on different amount of labeled documents. The results from the classification algorithms shows that MNB is better at classifying the data than LP.