Over the past decades, many industries have heavily spent on computerizing their work environments with the intention to simplify and expedite access to information and its processing. Typical of real-world data are various types of imperfections, uncertainties, ambiguities, that have complicated attempts at automated knowledge discovery. Indeed, it soon became obvious that adequate methods to deal with these problems were critically needed. Simple methods such as "interpolating" or just ignoring data imperfections being found often to lead to inferences of dubious practical value, the search for appropriate modification of knowledge-induction techniques began. Sometimes, rather non-standard approaches turned out to be necessary. For instance, the probabilistic approaches by earlier works are not sufficiently capable of handling the wider range of data imperfections that appear in many new applications (e.g., medical data). Dempster-Shafer theory provides a much stronger framework, and this is why it has been chosen as the fundamental paradigm exploited in this dissertation. The task of association rule mining is to detect frequently co-occurring groups of items in transactional databases. The majority of the papers in this field concentrate on how to expedite the search. Less attention has been devoted to how to employ the identified frequent itemsets for prediction purposes; worse still, methods to tailor association-mining techniques so that they can handle data imperfections are virtually nonexistent. This dissertation proposes a technique referred to by the acronym DS-ARM (Dempster-Shafer based Association Rule Mining) where the DS-theoretic framework is used to enhance a more traditional association-mining mechanism. Of particular interest is here a method to employ the knowledge of partial contents of a "shopping cart" for the prediction of what else the customer is likely to add to it. This formalized problem has many applications in the analysis of medical databases. A recently-proposed data structure, an itemset tree (IT-tree), is used to extract association rules in a computationally efficient manner, thus addressing the scalability problem that has disqualified more traditional techniques from real-world applications. The proposed algorithm is based on the Dempster-Shafer theory of evidence combination. Extensive experiments explore the algorithm's behavior; some of them use synthetically generated data, others relied on data obtained from a machine-learning repository, yet others use a movie ratings dataset or a HIV/AIDS patient dataset.

The present work develops several methodologies for solving engineering analysis and design problems involving uncertainties and evidences from multiple sources. The influence of uncertainties on the safety/failure of the system and on the warranty costs (to the manufacturer) are also investigated. Both single and multiple objective optimization problems are considered. A methodology is developed to combine the evidences available from single or multiple sources in the presence (or absence) of credibility information of the sources using modified Dempster Shafer Theory (DST) and Fuzzy Theory in the design of uncertain engineering systems. To optimally design a system, multiple objectives, such as to maximize the belief for the overall safety of the system, minimize the deflection, maximize the natural frequency and minimize the weight of an engineering structure under both deterministic and uncertain parameters, and subjected to multiple constraints are considered. We also study the various combination rules like Dempster's rule, Yager's rule, Inagaki's extreme rule, Zhang's center combination rule and Murphy's average combination rule for combining evidences from multiple sources. These rules are compared and a selection procedure was developed to assist the analyst in selecting the most suitable combination rule to combine various evidences obtained from multiple sources based on the nature of evidence sets. A weighted Dempster Shafer theory for interval-valued data (WDSTI) and weighted fuzzy theory for intervals (WFTI) were proposed for combining evidence when different credibilities are associated with the various sources of evidence. For solving optimization problems which cannot be solved using traditional gradient-based methods (such as those involving nonconvex functions and discontinuities), a modified Particle Swarm Optimization (PSO) algorithm is developed to include dynamic maximum velocity function and bounce method to solve both deterministic multi-objective problems and uncertain multi-objective problems (vertex method is used in addition to the modified PSO algorithm for uncertain parameters). A modified game theory approach (MGT) is coupled with the modified PSO algorithm to solve multi-objective optimization problems. In case of problems with multiple evidences, belief is calculated for a safe design (satisfying all constraints) using the vertex method and the modified PSO algorithm is used to solve the multi-objective optimization problems. The multiobjective problem related to the design of a composite laminate simply supported beam with center load is also considered to minimize the weight and maximize buckling load using modified game theory. A comparison of different warranty policies for both repairable and non repairable products and an automobile warranty optimization problem is considered to minimize the total warranty cost of the automobile with a constraint on the total failure probability of the system. To illustrate the methodologies presented in this work, several numerical design examples are solved. We finally present the conclusions along with a brief discussion of the future scope of the research.

O reconhecimento de atividaes é vital no contexto dos ambientes inteligentes. Mesmo com a facilidade de acesso a sensores móveis baratos, reconhecer atividades continua sendo um problema difícil devido à incerteza nas leituras dos sensores e à complexidade das atividades. A teoria da evidência provê um modelo de reconhecimento de atividades que detecta atividades mesmo na presença de incerteza nas leituras dos sensores, mas ainda não é capaz de modelar atividades complexas ou mudanças na configuração dos sensores ou do ambiente. Este trabalho propõe combinar abordagens baseadas em modelagem de conhecimento com a teoria da evidência, melhorando assim a construção dos modelos da última trazendo a reusabilidade, flexibilidade e semântica rica da primeira. / Activity recognition is a vital need in the field of ambient intelligence. It is essential for many internet of things applications including energy management, healthcare systems and home automation. But, even with the many cheap mobile sensors envisioned by the internet of things, activity recognition remains a hard problem. This is due to uncertainty in sensor readings and the complexity of activities themselves. Evidence theory models provide activity recognition even in the presence of uncertain sensor readings, but cannot yet model complex activities or dynamic changes in sensor and environment configurations. This work proposes combining knowledge-based approaches with evidence theory, improving the construction of evidence theory models for activity recognition by bringing reusability, flexibility and rich semantics.

O reconhecimento de atividaes é vital no contexto dos ambientes inteligentes. Mesmo com a facilidade de acesso a sensores móveis baratos, reconhecer atividades continua sendo um problema difícil devido à incerteza nas leituras dos sensores e à complexidade das atividades. A teoria da evidência provê um modelo de reconhecimento de atividades que detecta atividades mesmo na presença de incerteza nas leituras dos sensores, mas ainda não é capaz de modelar atividades complexas ou mudanças na configuração dos sensores ou do ambiente. Este trabalho propõe combinar abordagens baseadas em modelagem de conhecimento com a teoria da evidência, melhorando assim a construção dos modelos da última trazendo a reusabilidade, flexibilidade e semântica rica da primeira. / Activity recognition is a vital need in the field of ambient intelligence. It is essential for many internet of things applications including energy management, healthcare systems and home automation. But, even with the many cheap mobile sensors envisioned by the internet of things, activity recognition remains a hard problem. This is due to uncertainty in sensor readings and the complexity of activities themselves. Evidence theory models provide activity recognition even in the presence of uncertain sensor readings, but cannot yet model complex activities or dynamic changes in sensor and environment configurations. This work proposes combining knowledge-based approaches with evidence theory, improving the construction of evidence theory models for activity recognition by bringing reusability, flexibility and rich semantics.

O reconhecimento de atividaes é vital no contexto dos ambientes inteligentes. Mesmo com a facilidade de acesso a sensores móveis baratos, reconhecer atividades continua sendo um problema difícil devido à incerteza nas leituras dos sensores e à complexidade das atividades. A teoria da evidência provê um modelo de reconhecimento de atividades que detecta atividades mesmo na presença de incerteza nas leituras dos sensores, mas ainda não é capaz de modelar atividades complexas ou mudanças na configuração dos sensores ou do ambiente. Este trabalho propõe combinar abordagens baseadas em modelagem de conhecimento com a teoria da evidência, melhorando assim a construção dos modelos da última trazendo a reusabilidade, flexibilidade e semântica rica da primeira. / Activity recognition is a vital need in the field of ambient intelligence. It is essential for many internet of things applications including energy management, healthcare systems and home automation. But, even with the many cheap mobile sensors envisioned by the internet of things, activity recognition remains a hard problem. This is due to uncertainty in sensor readings and the complexity of activities themselves. Evidence theory models provide activity recognition even in the presence of uncertain sensor readings, but cannot yet model complex activities or dynamic changes in sensor and environment configurations. This work proposes combining knowledge-based approaches with evidence theory, improving the construction of evidence theory models for activity recognition by bringing reusability, flexibility and rich semantics.

Made available in DSpace on 2014-06-12T15:54:40Z (GMT). No. of bitstreams: 1 license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2005 / O presente trabalho estabelece uma extensão à Teoria Matemática da Evidência, também conhecida como Teoria de Dempster-Shafer, através da adoção de uma nova regra de combinação de evidências e de um arcabouço conceitual associado. Essa extensão resolve os problemas de comportamento contra-intuitivo apresentados originalmente pela teoria, amplia o poder expressional da mesma e permite a representação da incerteza nos resultados. A representação da incerteza implica a disponibilidade da mesma como um recurso estratégico a ser utilizado nas decisões baseadas nas evidências combinadas, bem como deixa explícita a relação entre os resultados numéricos obtidos e a probabilidade clássica

Made available in DSpace on 2014-06-12T15:55:07Z (GMT). No. of bitstreams: 2 arquivo9565_1.pdf: 714901 bytes, checksum: e1602dd43ff228b49b6ff591a1e8915a (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2005 / O presente trabalho estabelece uma extensão `a Teoria Matemática da Evidência, também conhecida como Teoria de Dempster-Shafer, através da adoção de uma nova regra de combinação de evidências e de um arcabouço conceitual associado. Essa extensão resolve os problemas de comportamento contra-intuitivo apresentados originalmente pela teoria, amplia o poder expressional da mesma e permite a representação da incerteza nos resultados. A representação da incerteza implica a disponibilidade da mesma como um recurso estratégico a ser utilizado nas decisões baseadas nas evidências combinadas, bem como deixa explícita a relação entre os resultados numéricos obtidos e a probabilidade clássica

This research explores the theory and practice of risk assessment and project evaluationand proposes novel alternatives. Reviewing literature revealed a continuous endeavourfor better project risk modelling and analysis. A number of proposals for improving theprevailing Probability-Impact (P-I) risk model can be found in literature. Moreover,researchers have investigated the feasibility of different theories in analysing projectrisk. Furthermore, various decision support systems (DSSs) are available for aidingpractitioners in risk assessment and decision making. Unfortunately, they are sufferingfrom a low take-up. Instead, personal judgment and past experience are mainly used foranalysing risk and making decisions.In this research, a new risk model is proposed through extending the P-I risk model toinclude a third dimension: probability of impact materialisation. Such an extensionreflects the characteristics of a risk, its surrounding environment and the ability ofmitigating its impact. A new assessment methodology is devised. Dempster-ShaferTheory of Evidence (DST) is researched and presented as a novel alternative toProbability Theory (PT) and Fuzzy Sets Theory (FST) which dominate the literature ofproject risks analysis. A DST-based assessment methodology was developed forstructuring the personal experience and professional judgment of risk analysts andutilising them for risk analysis. Benefiting from the unique features of the EvidentialReasoning (ER) approach, the proposed methodology enables analysts to express theirevaluations in distributed forms, so that they can provide degrees of belief in apredefined set of assessment grades based on available information. This is a veryeffective way for tackling the problem of lack of information which is an inherentfeature of most projects during the tendering stage. It is the first time that such anapproach is ever used for handling construction risk assessment. Monetary equivalent isused as a common scale for measuring risk impact on various project success objectives,and the evidential reasoning (ER) algorithm is used as an assessment aggregation toolinstead of the simple averaging procedure which might not be appropriate in allsituations. A DST-based project evaluation framework was developed using projectrisks and benefits as evaluation attributes. Monetary equivalent was used also as acommon scale for measuring project risks and benefits and the ER algorithm as anaggregation tool.The viability of the proposed risk model, assessment methodology and projectevaluation framework was investigated through conducting interviews with constructionprofessionals and administering postal and online questionnaires. A decision supportsystem (DSS) was devised to facilitate the proposed approaches and to perform therequired calculations. The DSS was developed in light of the research findingsregarding the reasons of low take-up of the existing tools. Four validation case studieswere conducted. Senior managers in separate British construction companies tested thetool and found it useful, helpful and easy to use.It is concluded that the proposed risk model, risk assessment methodology and projectevaluation framework could be viable alternatives to the existing ones. Professionalexperience was modelled and utilised systematically for risk and benefit analysis. Thismay help closing the gap between theory and practice of risk analysis and decisionmaking in construction. The research findings recommend further exploration of thepotential applications of DST and ER in construction management domain.

Self-driving cars have recently become a challenging research topic, with the aim of making transportation safer and more efficient. Current advanced driving assistance systems (ADAS) allow cars to drive autonomously by following lane markings, identifying road signs and detecting pedestrians and other vehicles. In this thesis work we improve the robustness of autonomous cars by designing an on-road obstacle detection system. The proposed solution consists on the low-level fusion of radar and lidar through the occupancy grid framework. Two inference theories are implemented and evaluated: Bayesian probability theory and Dempster-Shafer theory of evidence. Obstacle detection is performed through image processing of the occupancy grid. Last, the Dempster-Shafer additional features are leveraged by proposing a sensor performance estimation module and performing advanced conflict management. The work has been carried out at Volvo Car Corporation, where real experiments on a test vehicle have been performed under different environmental conditions and types of objects. The system has been evaluated according to the quality of the resulting occupancy grids, detection rate as well as information content in terms of entropy. The results show a significant improvement of the detection rate over single-sensor approaches. Furthermore, the Dempster-Shafer implementation may slightly outperform the Bayesian one when there is conflicting information, although the high computational cost limits its practical application. Last, we demonstrate that the proposed solution is easily scalable to include additional sensors.

While exploring an unknown environment, an intelligent agent has only its sensors to guide its actions. Each sensor's ability to provide accurate information depends on the environment's characteristics. If the agent does not know these characteristics, how can it determine which sensors to rely on? This problem is exacerbated by sensing anomalies: cases where sensor(s) are working but the readings lead to an incorrect interpretation of the environment, e.g. laser sensors cannot detect glass. This work addresses the following research question: Can an inconsistency-based sensing accuracy indicator, which relies solely on fused sensor readings, be used to detect and characterize sensing anomalies in unknown environments? A novel inconsistency-based approach was investigated for sensing anomaly detection and characterization by a mobile robot using range sensing for mapping. Based on the hypothesis that sensing anomalies manifest as inconsistent sensor readings, the approach employed Dempster-Shafer theory and six metrics from the evidential literature to measure the magnitude of inconsistency. These were applied directly to fused sensor data with a threshold, forming an indicator, used to distinguish minor noise from anomalous readings. Experiments with real sensor data from four indoor and two outdoor environments showed that three of the six evidential inconsistency metrics can partially address the issue of noticing sensing anomalies in unknown environments. Polaroid sonar sensors, SICK laser range finders, and a Canesta range camera were used. Despite extensive training in known environments, the indicators could not reliably detect sensing anomalies, i.e. distinguish them from ordinary noise. However, sensing accuracy could be estimated (correlations with sensor error exceeded 0.8) and regions with suspect readings could be isolated. Trained indicators failed to rank sensors, but improved map quality by resetting suspect regions (up to 57.65%) or guiding sensor selection (up to 75.86%). This work contributes to the robotics and uncertainty in artificial intelligence communities by establishing the use of evidential metrics for adapting a single sensor or identifying the most accurate sensor to optimize the sensing accuracy in unknown environments. Future applications could enable intelligent systems to switch information sources to optimize mission performance and identify the reliability of sources for different environments.