When triangulating a belief network we aim to obtain a junction tree of minimum state space. Searching for the optimal triangulation can be cast as a search over all the permutations of the network's vaeriables. Our approach is to embed the discrete set of permutations in a convex continuous domain D. By suitably extending the cost function over D and solving the continous nonlinear optimization task we hope to obtain a good triangulation with respect to the aformentioned cost. In this paper we introduce an upper bound to the total junction tree weight as the cost function. The appropriatedness of this choice is discussed and explored by simulations. Then we present two ways of embedding the new objective function into continuous domains and show that they perform well compared to the best known heuristic.

A Technical Due Diligence (TDD) investigation is an important step in the process of obtaining financing, or in mergers and acquisitions, for a wind power project. The investigation, the scope of which varies depending on the stage and nature of the project, involves reviewing important documentation relating to different aspects of the project, assessing potential risks in terms of the quality of the information available and suggesting mitigation or other risk management measures where required. A TDD assessment can greatly benefit from increased objectivity in terms of the reviewed aspects as it enables a sharper focus on the important risk elements and also provides a better appreciation of the investigated parameters. This master’s thesis has been an attempt to introduce more objectivity in the technical due diligence process followed at the host company. Thereafter, a points-based scoring system was devised to quantify the answered questions. The different aspects under investigation have a complex interrelationship and the resulting risks can be viewed as an outcome of a causal framework. To identify this causal framework the concept of Bayesian Belief Networks has been assessed. The resulting Bayesian Networks can be considered to provide a holistic framework for risk analysis within the TDD assessment process. The importance of accurate analysis of likelihood information for accurate analysis of Bayesian analysis has been identified. The statistical data set for the right framework needs to be generated to have the right correct setting for Bayesian analysis in the future studies. The objectiveness of the TDD process can be further enhanced by taking into consideration the capability of the investing body to handle the identified risks and also benchmarking risky aspects with industry standards or historical precedence.

China has continued the economic reform and open door policy over 30 years with many great achievements, such as the second largest GDP, the largest import and export economy with the largest infrastructural investment in the world. On the other hand, the conflicts and risks the firms especially for small and medium sized manufacturing enterprises (SMEs) have faced are extremely serious and more acute due to the economy growth and increasing social wealth, especially in Yangtze River Delta, in the general context of ever increasing cost such as labour, land and higher customers’ expectations such as the quality of product. These serious problems are challenges for the competitiveness of SMEs in Yangtze River Delta. This research aims to investigate and improve the competitiveness of SMEs by the main variables such as enterprise’s resources, product’s competitive issues and innovation activities related barriers. To achieve the aim, the research employed a mixed method of quantitative and qualitative approaches to build the competitiveness’s belief network model by Bayesian Belief Networks and analyze the factors of the most important variables by the SPSS software. Secondly, 36 entrepreneurs of small and medium sized manufacturing enterprises in Yangtze River Delta have been carefully selected to participate in the questionnaire survey and face to face interviews. All participants are entrepreneurs who have run enterprise for at least three years. Five kinds of resources, competitive issues and innovation have been identified as the variables of competitiveness. The findings of research are mainly related to the three aspects which are general view of variables; barriers to innovation activity and importance of variables for improving the competitiveness; and the factor analysis of quality management practices. Firstly, the general condition of financial resource is the worst in resource sector of SMEs; Dependability is the best performance in competitive issues of SMEs; Lack of finance is generally identified the biggest barrier to innovation of SMEs. Secondly, the Physical resource in resource sector and Quality in competitive issues sector are the most important variables for improving the competitiveness of SMEs after BBN assessment; Lack of technical experts is the most serious barrier when the SMEs are really focusing on the innovation according to the BBN assessments. Thirdly, the factor analyses have identified the key independent factors explaining the quality management practices in these SMEs. Finally, these findings can help the SMEs build variables’ impact tables based on the outputs from the conditional assessment of BBNs to make more efficient and effective decisions when they try to improve the enterprise competitiveness, with detailed recommendations. At the same time, the importance and factors of good quality management practices have also been argued to help the entrepreneurs improve the quality performance and their enterprise competitiveness.

Lean Six Sigma is a method used to improve the quality and efficiency of processes by reducing variation and eliminating wastes (non-value added activities) in an organisation. The concept of combining the principles and tools of Lean Enterprise and Six Sigma has been discussed in the literature. The majority of Lean Six Sigma applications in private industry have focused primarily on manufacturing applications. The literature has not provided a framework for implementing Lean Six Sigma programmes in non-manufacturing or transactional processes like those in the Entertainment Media industry. The Saudi Broadcasting Corporation (SBC), like many other industries in Saudi Arabia, has high occupational safety risks, such as electric, fire and fall hazards which often occur in the media workplace. These risks are considered very costly and affect productivity and employee morale in general. The main objective of this research is to provide a synergistic approach to integrating occupational health and safety programmes and Lean Six Sigma tools using the DMAIC (Define-Measure-Analyse-Improve-Control) problem-solving method to strengthen and assure the success of safety programmes in the Saudi Broadcasting Corporation (SBC). This research identifies the roadmap (i.e. activities, principles, tools, and important component factors) for applying Lean Six Sigma tools in the media industry. A case study addressing the safety issues that affect employees’ performance within the Saudi Broadcasting Corporation (SBC) TV studio is used to validate work outlined in this research. Furthermore, the Bayesian Belief Networks (BBN) method is used to understand the probability occurrence of safety hazards. The application of the Taguchi Experimental Design method and other Lean Six Sigma tools, such as Cause and Effect diagrams, Pareto principles, 5S, Value Stream map, and Poka-Yoke have been incorporated in to this research. The application of Lean Six Sigma DMAIC problem-solving tools resulted in significant improvement in safety within SBC. The average electrical hazard incident decreased from 2.08 to 0.33, the average fire hazard incident decreased from 1.25 to 0.08, and the average fall hazard incident decreased from 3.42 to 0.17. The research has important implications for the company and its employees, with positive outcomes and feedback reported by top management, the senior technicians, and experts. The research improved the safety by reducing electrical, fire and fall risks. The Safety training sessions are one of the most significant factors that improve their safety awareness. It is observed that Lean Six Sigma problem-solving tools and methods are effective in the Saudi Broadcasting Corporation (SBC).

Submitted by Maicon Juliano Schmidt (maicons) on 2015-06-12T19:13:14Z No. of bitstreams: 1 Ingo Jost.pdf: 1217467 bytes, checksum: bf67cd6724b1cd182a12a3cd7b5af1eb (MD5) / Made available in DSpace on 2015-06-12T19:13:14Z (GMT). No. of bitstreams: 1 Ingo Jost.pdf: 1217467 bytes, checksum: bf67cd6724b1cd182a12a3cd7b5af1eb (MD5) Previous issue date: 2015-02-26 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Deep Learning é uma sub-área de Aprendizado de Máquina que tem obtido resultados sa- tisfatórios em várias áreas de aplicação, implementada por diferentes algoritmos, como Stacked Auto-encoders ou Deep Belief Networks. Este trabalho propõe uma modelagem que aplica uma implementação de um classificador que aborda técnicas de Deep Learning em Mineração de Opiniões, área que tem sido alvo de constantes estudos, dada a necessidade das corporações buscarem a compreensão que clientes possuem de seus produtos ou serviços. O favorecimento do crescimento de Mineração de Opiniões também se dá pelo ambiente colaborativo da Web 2.0, em que várias ferramentas propiciam a emissão de opiniões. Os dados utilizados passaram por um refinamento na etapa de pré-processamento com o intuito de aplicar Deep Learning, da qual uma das principais atribuições é a seleção de características, em dados refinados em vez de dados mais brutos. A promissora tecnologia de Deep Learning combinada com a estratégia de refinamento demonstrou nos experimentos a obtenção de resultados competitivos com outros estudos relacionados e abrem perspectiva de extensão deste trabalho. / Deep Learning is a Machine Learning’s sub-area that have achieved satisfactory results in different application areas, implemented by different algorithms, such as Stacked Auto- encoders or Deep Belief Networks. This work proposes a research that applies a classifier that implements Deep Learning concepts in Opinion Mining, area has been approached by con- stant researches, due the need of corporations seeking the understanding that customers have of your products or services. The Opinion Mining’s growth is favored also by the collaborative Web 2.0 environment, where multiple tools provide issuing opinions. The data used for exper- iments were refined in preprocessing step in order to apply Deep Learning, which it one of the main tasks the feature selection, in refined data, instead of applying Deep Learning in more raw data. The refinement strategy combined with the promising technology of Deep Learning has demonstrated in preliminary experiments the achievement of competitive results with other studies and opens the perspective for extension of this work.

Como a complexidade das tarefas realizadas por robôs móveis vêm aumentando a cada dia, a percepção do robô deve ser capaz de capturar informações mais ricas do ambiente, que permitam a tomada de decisões complexas. Entre os possíveis tipos de informação que podem ser obtidos do ambiente, as informações geométricas e semânticas têm papéis importantes na maioria das tarefas designadas a robôs. Enquanto as informações geométricas revelam como os objetos e obstáculos estão distribuídos no espaço, as informações semânticas capturam a presença de estruturas complexas e eventos em andamento no ambiente, e os condensam em descrições abstratas. Esta tese propõe uma nova técnica probabilística para construir uma representação do ambiente baseada em objetos a partir de imagens capturadas por um robô navegando com uma câmera de vídeo solidária a ele. Esta representação, que fornece descrições geométricas e semânticas de objetos, é chamada O-Map, e é a primeira do gênero no contexto de navegação de robôs. A técnica de mapeamento proposta é também nova, e resolve concomitantemente os problemas de localização, mapeamento e classificação de objetos, que surgem quando da construção de O-Maps usando imagens processadas por detectores imperfeitos de objetos e sem um sensor de localização global. Por este motivo, a técnica proposta é chamada O-SLAM, e é o primeiro algoritmo que soluciona simultaneamente os problemas de localização e mapeamento usando somente odometria e o resultado de algoritmos de reconhecimento de objetos. Os resultados obtidos através da aplicação de O-SLAM em imagens processadas por uma técnica simples de detecção de objetos mostra que o algoritmo proposto é capaz de construir mapas que descrevem consistentemente os objetos do ambiente, dado que o sistema de visão computacional seja capaz de detectá-los regularmente. Em particular, O-SLAM é eficaz em fechar voltas grandes na trajetória do robô, e obtém sucesso mesmo se o sistema de detecção de objetos posuir falhas, relatando falsos positivos e errando a classe do objeto algumas vezes, consertando estes erros. Dessa forma, O-SLAM é um passo em direção à solução integrada do problema de localização, mapeamento e reconhecimento de objetos, a qual deve prescindir de um sistema pronto de reconhecimento de objetos e gerar O-Maps somente pela fusão de informações geométricas e visuais obtidas pelo robô. / As tasks performed by mobile robots are increasing in complexity, robot perception must be able to capture richer information from the environment in order to allow complex decision making. Among the possible types of information that can be retrieved from the environment, geometric and semantic information play important roles in most of the tasks assigned to robots. While geometric information reveals how objects and obstacles are distributed in space, semantic information captures the presence of complex structures and ongoing events from the environment and summarize them in abstract descriptions. This thesis proposes a new probabilistic technique to build an object-based representation of the robot surrounding environment using images captured by an attached video camera. This representation, which provides geometric and semantic descriptions of the objects, is called O-Map, and is the first of its kind in the robot navigation context. The proposed mapping technique is also new, and concurrently solves the localization, mapping and object classification problems arisen from building O-Maps using images processed by imperfect object detectors and no global localization sensor. Thus, the proposed technique is called O-SLAM, and is the first algorithm to solve the simultaneous localization and mapping problem using solely odometers and the output from object recognition algorithms. The results obtained by applying O-SLAM to images processed by simple a object detection technique show that the proposed algorithm is able to build consistent maps describing the objects in the environment, provided that the computer vision system is able to detect them on a regular basis. In particular, O-SLAM is effective in closing large loops in the trajectory, and is able to perform well even if the object detection system makes spurious detections and reports wrong object classes, fixing these errors. Thus, O-SLAM is a step towards the solution of the simultaneous localization, mapping and object recognition problem, which must drop the need for an off-the-shelf object recognition system and generate O-Maps only by fusing geometric and appearance information gathered by the robot.

Como a complexidade das tarefas realizadas por robôs móveis vêm aumentando a cada dia, a percepção do robô deve ser capaz de capturar informações mais ricas do ambiente, que permitam a tomada de decisões complexas. Entre os possíveis tipos de informação que podem ser obtidos do ambiente, as informações geométricas e semânticas têm papéis importantes na maioria das tarefas designadas a robôs. Enquanto as informações geométricas revelam como os objetos e obstáculos estão distribuídos no espaço, as informações semânticas capturam a presença de estruturas complexas e eventos em andamento no ambiente, e os condensam em descrições abstratas. Esta tese propõe uma nova técnica probabilística para construir uma representação do ambiente baseada em objetos a partir de imagens capturadas por um robô navegando com uma câmera de vídeo solidária a ele. Esta representação, que fornece descrições geométricas e semânticas de objetos, é chamada O-Map, e é a primeira do gênero no contexto de navegação de robôs. A técnica de mapeamento proposta é também nova, e resolve concomitantemente os problemas de localização, mapeamento e classificação de objetos, que surgem quando da construção de O-Maps usando imagens processadas por detectores imperfeitos de objetos e sem um sensor de localização global. Por este motivo, a técnica proposta é chamada O-SLAM, e é o primeiro algoritmo que soluciona simultaneamente os problemas de localização e mapeamento usando somente odometria e o resultado de algoritmos de reconhecimento de objetos. Os resultados obtidos através da aplicação de O-SLAM em imagens processadas por uma técnica simples de detecção de objetos mostra que o algoritmo proposto é capaz de construir mapas que descrevem consistentemente os objetos do ambiente, dado que o sistema de visão computacional seja capaz de detectá-los regularmente. Em particular, O-SLAM é eficaz em fechar voltas grandes na trajetória do robô, e obtém sucesso mesmo se o sistema de detecção de objetos posuir falhas, relatando falsos positivos e errando a classe do objeto algumas vezes, consertando estes erros. Dessa forma, O-SLAM é um passo em direção à solução integrada do problema de localização, mapeamento e reconhecimento de objetos, a qual deve prescindir de um sistema pronto de reconhecimento de objetos e gerar O-Maps somente pela fusão de informações geométricas e visuais obtidas pelo robô. / As tasks performed by mobile robots are increasing in complexity, robot perception must be able to capture richer information from the environment in order to allow complex decision making. Among the possible types of information that can be retrieved from the environment, geometric and semantic information play important roles in most of the tasks assigned to robots. While geometric information reveals how objects and obstacles are distributed in space, semantic information captures the presence of complex structures and ongoing events from the environment and summarize them in abstract descriptions. This thesis proposes a new probabilistic technique to build an object-based representation of the robot surrounding environment using images captured by an attached video camera. This representation, which provides geometric and semantic descriptions of the objects, is called O-Map, and is the first of its kind in the robot navigation context. The proposed mapping technique is also new, and concurrently solves the localization, mapping and object classification problems arisen from building O-Maps using images processed by imperfect object detectors and no global localization sensor. Thus, the proposed technique is called O-SLAM, and is the first algorithm to solve the simultaneous localization and mapping problem using solely odometers and the output from object recognition algorithms. The results obtained by applying O-SLAM to images processed by simple a object detection technique show that the proposed algorithm is able to build consistent maps describing the objects in the environment, provided that the computer vision system is able to detect them on a regular basis. In particular, O-SLAM is effective in closing large loops in the trajectory, and is able to perform well even if the object detection system makes spurious detections and reports wrong object classes, fixing these errors. Thus, O-SLAM is a step towards the solution of the simultaneous localization, mapping and object recognition problem, which must drop the need for an off-the-shelf object recognition system and generate O-Maps only by fusing geometric and appearance information gathered by the robot.

Companies, across the globe are concerned with risks that impair their ability to produce quality products at a low cost and deliver them to customers on time. Risk assessment, comprising of both external and internal elements, prepares companies to identify and manage the risks affecting them. Although both external/supply chain and internal/production line risk assessments are necessary, internal risk assessment is often ignored. Internal risk assessment helps companies recognize vulnerable sections of production operations and provide opportunities for risk mitigation. In this research, a novel production line risk assessment methodology is proposed. Traditional simulation techniques fail to capture the complex relationship amongst risk events and the dynamic interaction between risks affecting a production line. Bayesian- integrated System Dynamics modelling can help resolve this limitation. Bayesian Belief Networks (BBN) effectively capture risk relationships and their likelihoods. Integrating BBN with System Dynamics (SD) for modelling production lines help capture the impact of risk events on a production line as well as the dynamic interaction between those risks and production line variables. The proposed methodology is applied to an industrial case study for validation and to discern research and practical implications.

Sigmoid type belief networks, a class of probabilistic neural networks, provide a natural framework for compactly representing probabilistic information in a variety of unsupervised and supervised learning problems. Often the parameters used in these networks need to be learned from examples. Unfortunately, estimating the parameters via exact probabilistic calculations (i.e, the EM-algorithm) is intractable even for networks with fairly small numbers of hidden units. We propose to avoid the infeasibility of the E step by bounding likelihoods instead of computing them exactly. We introduce extended and complementary representations for these networks and show that the estimation of the network parameters can be made fast (reduced to quadratic optimization) by performing the estimation in either of the alternative domains. The complementary networks can be used for continuous density estimation as well.

Καθημερινά στα εξωτερικά ιατρεία των νοσοκομείων μας αντιμετωπίζονται πάρα πολλά περιστατικά ελαφρών κρανιοεγκεφαλικων κακώσεων. Πολλά από αυτά ακολουθούνται από παράπονα για εμφάνιση μετατραυματικων συμπτωμάτων όπως ζαλάδες ή και μερικές φορές πιο σοβαρών, το σύνολο των οποίων αποτελούν το μεταδιασεισικό σύνδρομο. Ο συνδυασμός των κρανιοεγκεφαλικών κακώσεων και του μεταδιασεισικού συνδρόμου αποτελούν σοβαρό πρόβλημα στις σύγχρονες κοινωνίες γιατί είναι η αιτία για την επαγγελματική ανικανότητα των πασχόντων αποτελώντας έτσι σοβαρό οικονομικό αντίκτυπο. Σκοπός της παρούσας μελέτης είναι η οικοδόμηση ενός εργαλείου υποστήριξης ιατρικών αποφάσεων που θα μπορεί να εκτιμήσει ποσοτικά την πιθανότητα εμφάνισης του μεταδιασεισικού συνδρόμου σε κάποιον ασθενή με ήπια κρανιοεγκεφαλική κάκωση, στηριζόμενο σε προγνωστικούς παράγοντες που αναδείχθηκαν από δεδομένα που συγκεντρώθηκαν στις εξετάσεις που πραγματοποιούνται στα εξωτερικά ιατρεία. Υλικό-Μέθοδοι: Η μελέτη πραγματοποιήθηκε στο Πανεπιστημιακό Γενικό Νοσοκομείο Πατρών, στη Δυτική Ελλάδα. Η συλλογή των μετρήσεων έγινε στα πλαίσια διδακτορικής διατριβής της Νευροχειρουργικής Κλινικής του Τμήματος Ιατρικής Οι μετρήσεις αυτές χρησιμοποιήθηκαν στη παρούσα μελέτη για την εξαγωγή των δικών μας αποτελεσμάτων. Συνολικά καταγράφηκαν μετρήσεις από 539 ασθενείς με ήπια κρανιοεγκεφαλική κάκωση. 223 από αυτούς τους ασθενείς πληρούσαν τα κριτήρια του "Colorado Medical Society Guidelines" για τον καθορισμό της διάσεισης, με μέση ηλικία τα 30 έτη (εύρος: 18.5-57.5). Για την εξαγωγή των αποτελεσμάτων κατασκευάστηκε ένα δίκτυο πεποίθησης και εντάχθηκε στο λογισμικό Netica για την μαθηματική του ανάλυση. Συμπεράσματα: Μετά την εξεργασία των μετρήσεων πρόεκυψαν ενδιαφέροντα αποτελέσματα σχετικά με την πρόβλεψη εμφάνισης του μεταδιασεισικου συνδρόμου. Φάνηκε λοιπόν ότι δυο από τους πλέον σημαντικούς παράγοντες είναι το φύλο και το είδος του ατυχήματος εξαιτίας του οποίου υπέστη ο ασθενής την κρανιοεγκεφαλικη κάκωση. Χαρακτηριστικά μπορούμε να δούμε ότι η πιθανότητα εμφάνισης για μια γυναίκα που ενεπλακη σε αυτοκινητιστικό δυστύχημα είναι 42,7%, ενώ αντίστοιχα για έναν άντρα είναι 19,8%. Επίσης σε μια γυναίκα που δέχτηκε επίθεση η πιθανότητα είναι 37,4%, ενώ για έναν άντρα είναι 16,5%. Μέσω της χρήσης του Δικτύου Πεποίθησης μπορούμε να ξέρουμε από πριν ποιες είναι εκείνες οι μεταβλητές που με την παρουσία τους αυξάνουν ή ελαττώνουν την πιθανότητα εμφάνισης του μεταδιασεισικού συνδρόμου και να ποσοτικοποιήσουμε αυτή την επίδραση. / Outpatient departments in hospitals treat many cases of mild traumatic brain injury daily. Many of these complaints are followed by the appearance of post traumatic symptoms as dizziness or sometimes more serious, which comprise the post-concussion syndrome. The combination of craniocerebral injuries and post-concussion syndrome is a serious problem in modern societies because it is the cause of occupational disability of patients, thus constituting a serious economic impact. The purpose of this study is to combine the main prognostic factors that lead to post-concussion syndrome in an adult Greek population after a traumatic brain injury into a decision support tool that could be useful in the outpatient department. Material-Methods: The study was conducted at the University Hospital of Patras, Western Greece. The collection of the measurements which were used in this study were made in the context of a doctoral thesis of the Neurosurgery Department. Overall, 539 patients with mild head-injuries were recorded. Of these, 223 patients patients met the criteria of "Colorado Medical Society Guidelines" to determine concussion, with a mean age of 30 years (range: 18.5-57.5). Based on these data, a naïve Bayesian Network was constructed and Netica software for the mathematical analysis that followed. Results: The use of the Bayesian Network allows us to measure the impact of certain prognostic factors to the probability of occurrence of post-concussion syndrome. It was found, that two of the most important factors is the gender and the type of accident which the patient suffered a traumatic brain injury. As an example, the estimated probability to develop a post-concussion syndrome for a woman who was involved in a car accident is 42.7%, while for a man is 19.8%. Also, for a woman who was attacked, the probability is 37.4%, while for a man is 16.5%.