983 resultados para bayesian networks
Resumo:
Unlike the evaluation of single items of scientific evidence, the formal study and analysis of the jointevaluation of several distinct items of forensic evidence has to date received some punctual, ratherthan systematic, attention. Questions about the (i) relationships among a set of (usually unobservable)propositions and a set of (observable) items of scientific evidence, (ii) the joint probative valueof a collection of distinct items of evidence as well as (iii) the contribution of each individual itemwithin a given group of pieces of evidence still represent fundamental areas of research. To somedegree, this is remarkable since both, forensic science theory and practice, yet many daily inferencetasks, require the consideration of multiple items if not masses of evidence. A recurrent and particularcomplication that arises in such settings is that the application of probability theory, i.e. the referencemethod for reasoning under uncertainty, becomes increasingly demanding. The present paper takesthis as a starting point and discusses graphical probability models, i.e. Bayesian networks, as frameworkwithin which the joint evaluation of scientific evidence can be approached in some viable way.Based on a review of existing main contributions in this area, the article here aims at presentinginstances of real case studies from the author's institution in order to point out the usefulness andcapacities of Bayesian networks for the probabilistic assessment of the probative value of multipleand interrelated items of evidence. A main emphasis is placed on underlying general patterns of inference,their representation as well as their graphical probabilistic analysis. Attention is also drawnto inferential interactions, such as redundancy, synergy and directional change. These distinguish thejoint evaluation of evidence from assessments of isolated items of evidence. Together, these topicspresent aspects of interest to both, domain experts and recipients of expert information, because theyhave bearing on how multiple items of evidence are meaningfully and appropriately set into context.
Resumo:
Due to the rise of criminal, civil and administrative judicial situations involving people lacking valid identity documents, age estimation of living persons has become an important operational procedure for numerous forensic and medicolegal services worldwide. The chronological age of a given person is generally estimated from the observed degree of maturity of some selected physical attributes by means of statistical methods. However, their application in the forensic framework suffers from some conceptual and practical drawbacks, as recently claimed in the specialised literature. The aim of this paper is therefore to offer an alternative solution for overcoming these limits, by reiterating the utility of a probabilistic Bayesian approach for age estimation. This approach allows one to deal in a transparent way with the uncertainty surrounding the age estimation process and to produce all the relevant information in the form of posterior probability distribution about the chronological age of the person under investigation. Furthermore, this probability distribution can also be used for evaluating in a coherent way the possibility that the examined individual is younger or older than a given legal age threshold having a particular legal interest. The main novelty introduced by this work is the development of a probabilistic graphical model, i.e. a Bayesian network, for dealing with the problem at hand. The use of this kind of probabilistic tool can significantly facilitate the application of the proposed methodology: examples are presented based on data related to the ossification status of the medial clavicular epiphysis. The reliability and the advantages of this probabilistic tool are presented and discussed.
Resumo:
The evaluation of forensic evidence can occur at any level within the hierarchy of propositions depending on the question being asked and the amount and type of information that is taken into account within the evaluation. Commonly DNA evidence is reported given propositions that deal with the sub-source level in the hierarchy, which deals only with the possibility that a nominated individual is a source of DNA in a trace (or contributor to the DNA in the case of a mixed DNA trace). We explore the use of information obtained from examinations, presumptive and discriminating tests for body fluids, DNA concentrations and some case circumstances within a Bayesian network in order to provide assistance to the Courts that have to consider propositions at source level. We use a scenario in which the presence of blood is of interest as an exemplar and consider how DNA profiling results and the potential for laboratory error can be taken into account. We finish with examples of how the results of these reports could be presented in court using either numerical values or verbal descriptions of the results.
Resumo:
The problem of understanding how humans perceive the quality of a reproduced image is of interest to researchers of many fields related to vision science and engineering: optics and material physics, image processing (compression and transfer), printing and media technology, and psychology. A measure for visual quality cannot be defined without ambiguity because it is ultimately the subjective opinion of an end-user observing the product. The purpose of this thesis is to devise computational methods to estimate the overall visual quality of prints, i.e. a numerical value that combines all the relevant attributes of the perceived image quality. The problem is limited to consider the perceived quality of printed photographs from the viewpoint of a consumer, and moreover, the study focuses only on digital printing methods, such as inkjet and electrophotography. The main contributions of this thesis are two novel methods to estimate the overall visual quality of prints. In the first method, the quality is computed as a visible difference between the reproduced image and the original digital (reference) image, which is assumed to have an ideal quality. The second method utilises instrumental print quality measures, such as colour densities, measured from printed technical test fields, and connects the instrumental measures to the overall quality via subjective attributes, i.e. attributes that directly contribute to the perceived quality, using a Bayesian network. Both approaches were evaluated and verified with real data, and shown to predict well the subjective evaluation results.
Resumo:
The theme of this thesis is context-speci c independence in graphical models. Considering a system of stochastic variables it is often the case that the variables are dependent of each other. This can, for instance, be seen by measuring the covariance between a pair of variables. Using graphical models, it is possible to visualize the dependence structure found in a set of stochastic variables. Using ordinary graphical models, such as Markov networks, Bayesian networks, and Gaussian graphical models, the type of dependencies that can be modeled is limited to marginal and conditional (in)dependencies. The models introduced in this thesis enable the graphical representation of context-speci c independencies, i.e. conditional independencies that hold only in a subset of the outcome space of the conditioning variables. In the articles included in this thesis, we introduce several types of graphical models that can represent context-speci c independencies. Models for both discrete variables and continuous variables are considered. A wide range of properties are examined for the introduced models, including identi ability, robustness, scoring, and optimization. In one article, a predictive classi er which utilizes context-speci c independence models is introduced. This classi er clearly demonstrates the potential bene ts of the introduced models. The purpose of the material included in the thesis prior to the articles is to provide the basic theory needed to understand the articles.
Resumo:
Nous proposons une approche probabiliste afin de dterminer limpact des changements dans les programmes objets. Cette approche sert prdire, pour un changement donn dans une classe du systme, lensemble des autres classes potentiellement affectes par ce changement. Cette prdiction est donne sous la forme dune probabilit qui dpend dune part, des interactions entre les classes exprimes en termes de nombre dinvocations et dautre part, des relations extraites partir du code source. Ces relations sont extraites automatiquement par rtro-ingnierie. Pour la mise en oeuvre de notre approche, nous proposons une approche base sur les rseaux baysiens. Aprs une phase dapprentissage, ces rseaux prdisent lensemble des classes affectes par un changement. Lapproche probabiliste propose est value avec deux scnarios distincts mettant en oeuvre plusieurs types de changements effectus sur diffrents systmes. Pour les systmes qui possdent des donnes historiques, lapprentissage a t ralis partir des anciennes versions. Pour les systmes dont on ne possde pas assez de donnes relatives aux changements de ses versions antcdentes, lapprentissage a t ralis laide des donnes extraites dautres systmes.
Resumo:
La modlisation de lexprience de lutilisateur dans les Interactions Homme-Machine est un enjeu important pour la conception et le dveloppement des systmes adaptatifs intelligents. Dans ce contexte, une attention particulire est porte sur les ractions motionnelles de lutilisateur, car elles ont une influence capitale sur ses aptitudes cognitives, comme la perception et la prise de dcision. La modlisation des motions est particulirement pertinente pour les Systmes Tutoriels motionnellement Intelligents (STEI). Ces systmes cherchent identifier les motions de lapprenant lors des sessions dapprentissage, et optimiser son exprience dinteraction en recourant diverses stratgies dinterventions. Cette thse vise amliorer les mthodes de modlisation des motions et les stratgies motionnelles utilises actuellement par les STEI pour agir sur les motions de lapprenant. Plus prcisment, notre premier objectif a t de proposer une nouvelle mthode pour dtecter ltat motionnel de lapprenant, en utilisant diffrentes sources dinformations qui permettent de mesurer les motions de faon prcise, tout en tenant compte des variables individuelles qui peuvent avoir un impact sur la manifestation des motions. Pour ce faire, nous avons dvelopp une approche multimodale combinant plusieurs mesures physiologiques (activit crbrale, ractions galvaniques et rythme cardiaque) avec des variables individuelles, pour dtecter une motion trs frquemment observe lors des sessions dapprentissage, savoir lincertitude. Dans un premier lieu, nous avons identifi les indicateurs physiologiques cls qui sont associs cet tat, ainsi que les caractristiques individuelles qui contribuent sa manifestation. Puis, nous avons dvelopp des modles prdictifs permettant de dtecter automatiquement cet tat partir des diffrentes variables analyses, travers lentrainement dalgorithmes dapprentissage machine. Notre deuxime objectif a t de proposer une approche unifie pour reconnatre simultanment une combinaison de plusieurs motions, et valuer explicitement limpact de ces motions sur lexprience dinteraction de lapprenant. Pour cela, nous avons dvelopp une plateforme hirarchique, probabiliste et dynamique permettant de suivre les changements motionnels de l'apprenant au fil du temps, et dinfrer automatiquement la tendance gnrale qui caractrise son exprience dinteraction savoir : limmersion, le blocage ou le dcrochage. Limmersion correspond une exprience optimale : un tat dans lequel l'apprenant est compltement concentr et impliqu dans lactivit dapprentissage. Ltat de blocage correspond une tendance dinteraction non optimale o l'apprenant a de la difficult se concentrer. Finalement, le dcrochage correspond un tat extrmement dfavorable o lapprenant nest plus du tout impliqu dans lactivit dapprentissage. La plateforme propose intgre trois modalits de variables diagnostiques permettant dvaluer lexprience de lapprenant savoir : des variables physiologiques, des variables comportementales, et des mesures de performance, en combinaison avec des variables prdictives qui reprsentent le contexte courant de linteraction et les caractristiques personnelles de l'apprenant. Une tude a t ralise pour valider notre approche travers un protocole exprimental permettant de provoquer dlibrment les trois tendances cibles durant linteraction des apprenants avec diffrents environnements dapprentissage. Enfin, notre troisime objectif a t de proposer de nouvelles stratgies pour influencer positivement ltat motionnel de lapprenant, sans interrompre la dynamique de la session dapprentissage. Nous avons cette fin introduit le concept de stratgies motionnelles implicites : une nouvelle approche pour agir subtilement sur les motions de lapprenant, dans le but damliorer son exprience dapprentissage. Ces stratgies utilisent la perception subliminale, et plus prcisment une technique connue sous le nom damorage affectif. Cette technique permet de solliciter inconsciemment les motions de lapprenant, travers la projection damorces comportant certaines connotations affectives. Nous avons mis en uvre une stratgie motionnelle implicite utilisant une forme particulire damorage affectif savoir : le conditionnement valuatif, qui est destin amliorer de faon inconsciente lestime de soi. Une tude exprimentale a t ralise afin dvaluer limpact de cette stratgie sur les ractions motionnelles et les performances des apprenants.
Resumo:
We investigate solution sets of a special kind of linear inequality systems. In particular, we derive characterizations of these sets in terms of minimal solution sets. The studied inequalities emerge as information inequalities in the context of Bayesian networks. This allows to deduce important properties of Bayesian networks, which is important within causal inference.
Resumo:
The paper discusses maintenance challenges of organisations with a huge number of devices and proposes the use of probabilistic models to assist monitoring and maintenance planning. The proposal assumes connectivity of instruments to report relevant features for monitoring. Also, the existence of enough historical registers with diagnosed breakdowns is required to make probabilistic models reliable and useful for predictive maintenance strategies based on them. Regular Markov models based on estimated failure and repair rates are proposed to calculate the availability of the instruments and Dynamic Bayesian Networks are proposed to model cause-effect relationships to trigger predictive maintenance services based on the influence between observed features and previously documented diagnostics
Resumo:
Many weeds occur in patches but farmers frequently spray whole fields to control the weeds in these patches. Given a geo-referenced weed map, technology exists to confine spraying to these patches. Adoption of patch spraying by arable farmers has, however, been negligible partly due to the difficulty of constructing weed maps. Building on previous DEFRA and HGCA projects, this proposal aims to develop and evaluate a machine vision system to automate the weed mapping process. The project thereby addresses the principal technical stumbling block to widespread adoption of site specific weed management (SSWM). The accuracy of weed identification by machine vision based on a single field survey may be inadequate to create herbicide application maps. We therefore propose to test the hypothesis that sufficiently accurate weed maps can be constructed by integrating information from geo-referenced images captured automatically at different times of the year during normal field activities. Accuracy of identification will also be increased by utilising a priori knowledge of weeds present in fields. To prove this concept, images will be captured from arable fields on two farms and processed offline to identify and map the weeds, focussing especially on black-grass, wild oats, barren brome, couch grass and cleavers. As advocated by Lutman et al. (2002), the approach uncouples the weed mapping and treatment processes and builds on the observation that patches of these weeds are quite stable in arable fields. There are three main aspects to the project. 1) Machine vision hardware. Hardware component parts of the system are one or more cameras connected to a single board computer (Concurrent Solutions LLC) and interfaced with an accurate Global Positioning System (GPS) supplied by Patchwork Technology. The camera(s) will take separate measurements for each of the three primary colours of visible light (red, green and blue) in each pixel. The basic proof of concept can be achieved in principle using a single camera system, but in practice systems with more than one camera may need to be installed so that larger fractions of each field can be photographed. Hardware will be reviewed regularly during the project in response to feedback from other work packages and updated as required. 2) Image capture and weed identification software. The machine vision system will be attached to toolbars of farm machinery so that images can be collected during different field operations. Images will be captured at different ground speeds, in different directions and at different crop growth stages as well as in different crop backgrounds. Having captured geo-referenced images in the field, image analysis software will be developed to identify weed species by Murray State and Reading Universities with advice from The Arable Group. A wide range of pattern recognition and in particular Bayesian Networks will be used to advance the state of the art in machine vision-based weed identification and mapping. Weed identification algorithms used by others are inadequate for this project as we intend to collect and correlate images collected at different growth stages. Plants grown for this purpose by Herbiseed will be used in the first instance. In addition, our image capture and analysis system will include plant characteristics such as leaf shape, size, vein structure, colour and textural pattern, some of which are not detectable by other machine vision systems or are omitted by their algorithms. Using such a list of features observable using our machine vision system, we will determine those that can be used to distinguish weed species of interest. 3) Weed mapping. Geo-referenced maps of weeds in arable fields (Reading University and Syngenta) will be produced with advice from The Arable Group and Patchwork Technology. Natural infestations will be mapped in the fields but we will also introduce specimen plants in pots to facilitate more rigorous system evaluation and testing. Manual weed maps of the same fields will be generated by Reading University, Syngenta and Peter Lutman so that the accuracy of automated mapping can be assessed. The principal hypothesis and concept to be tested is that by combining maps from several surveys, a weed map with acceptable accuracy for endusers can be produced. If the concept is proved and can be commercialised, systems could be retrofitted at low cost onto existing farm machinery. The outputs of the weed mapping software would then link with the precision farming options already built into many commercial sprayers, allowing their use for targeted, site-specific herbicide applications. Immediate economic benefits would, therefore, arise directly from reducing herbicide costs. SSWM will also reduce the overall pesticide load on the crop and so may reduce pesticide residues in food and drinking water, and reduce adverse impacts of pesticides on non-target species and beneficials. Farmers may even choose to leave unsprayed some non-injurious, environmentally-beneficial, low density weed infestations. These benefits fit very well with the anticipated legislation emerging in the new EU Thematic Strategy for Pesticides which will encourage more targeted use of pesticides and greater uptake of Integrated Crop (Pest) Management approaches, and also with the requirements of the Water Framework Directive to reduce levels of pesticides in water bodies. The greater precision of weed management offered by SSWM is therefore a key element in preparing arable farming systems for the future, where policy makers and consumers want to minimise pesticide use and the carbon footprint of farming while maintaining food production and security. The mapping technology could also be used on organic farms to identify areas of fields needing mechanical weed control thereby reducing both carbon footprints and also damage to crops by, for example, spring tines. Objective i. To develop a prototype machine vision system for automated image capture during agricultural field operations; ii. To prove the concept that images captured by the machine vision system over a series of field operations can be processed to identify and geo-reference specific weeds in the field; iii. To generate weed maps from the geo-referenced, weed plants/patches identified in objective (ii).
Resumo:
A situation assessment uses reports from sensors to produce hypotheses about a situation at a level of aggregation that is of direct interest to a military commander. A low level of aggregation could mean forming tracks from reports, which is well documented in the tracking literature as track initiation and data association. In this paper there is also discussion on higher level aggregation; assessing the membership of tracks to larger groups. Ideas used in joint tracking and identification are extended, using multi-entity Bayesian networks to model a number of static variables, of which the identity of a target is one. For higher level aggregation a scheme for hypothesis management is required. It is shown how an offline clustering of vehicles can be reduced to an assignment problem.
Resumo:
The substitution of missing values, also called imputation, is an important data preparation task for many domains. Ideally, the substitution of missing values should not insert biases into the dataset. This aspect has been usually assessed by some measures of the prediction capability of imputation methods. Such measures assume the simulation of missing entries for some attributes whose values are actually known. These artificially missing values are imputed and then compared with the original values. Although this evaluation is useful, it does not allow the influence of imputed values in the ultimate modelling task (e.g. in classification) to be inferred. We argue that imputation cannot be properly evaluated apart from the modelling task. Thus, alternative approaches are needed. This article elaborates on the influence of imputed values in classification. In particular, a practical procedure for estimating the inserted bias is described. As an additional contribution, we have used such a procedure to empirically illustrate the performance of three imputation methods (majority, naive Bayes and Bayesian networks) in three datasets. Three classifiers (decision tree, naive Bayes and nearest neighbours) have been used as modelling tools in our experiments. The achieved results illustrate a variety of situations that can take place in the data preparation practice.
Resumo:
A crucial aspect of evidential reasoning in crime investigation involves comparing the support that evidence provides for alternative hypotheses. Recent work in forensic statistics has shown how Bayesian Networks (BNs) can be employed for this purpose. However, the specification of BNs requires conditional probability tables describing the uncertain processes under evaluation. When these processes are poorly understood, it is necessary to rely on subjective probabilities provided by experts. Accurate probabilities of this type are normally hard to acquire from experts. Recent work in qualitative reasoning has developed methods to perform probabilistic reasoning using coarser representations. However, the latter types of approaches are too imprecise to compare the likelihood of alternative hypotheses. This paper examines this shortcoming of the qualitative approaches when applied to the aforementioned problem, and identifies and integrates techniques to refine them.
Resumo:
O objetivo deste trabalho testar a aplicao de um modelo grfico probabilstico, denominado genericamente de Redes Bayesianas, para desenvolver modelos computacionais que possam ser utilizados para auxiliar a compreenso de problemas e/ou na previso de variveis de natureza econmica. Com este propsito, escolheu-se um problema amplamente abordado na literatura e comparou-se os resultados tericos e experimentais j consolidados com os obtidos utilizando a tcnica proposta. Para tanto,foi construdo um modelo para a classificao da tendncia do "risco pas" para o Brasil a partir de uma base de dados composta por variveis macroeconmicas e financeiras. Como medida do risco adotou-se o EMBI+ (Emerging Markets Bond Index Plus), por ser um indicador amplamente utilizado pelo mercado.
Resumo:
Durante os ltimos anos as reas de pesquisa sobre Agentes Inteligentes, Sistemas Multiagentes e Comunicao entre Agentes tm contribudo com uma revoluo na forma como sistemas inteligentes podem ser concebidos, fundamentados e construdos. Sendo assim, parece razovel supor que sistemas inteligentes que trabalhem com domnios probabilsticos de conhecimento possam compartilhar do mesmo tipo de benefcios que os sistemas mais tradicionais da Inteligncia Artificial receberam quando adotaram as concepes de agncia, de sistemas compostos de mltiplos agentes e de linguagens de comunicao entre estes agentes. Porm, existem dvidas no s sobre como se poderia escalar efetivamente um sistema probabilstico para uma arquitetura multiagente, mas como se poderia lidar com as questes relativas comunicao e representao de conhecimentos probabilsticos neste tipo de sistema, principalmente tendo em vista as limitaes das linguagens de comunicao entre agentes atuais, que no permitem comunicar ou representar este tipo de conhecimento. Este trabalho parte destas consideraes e prope uma generalizao do modelo terico puramente lgico que atualmente fundamenta a comunicao nos sistemas multiagentes, que ser capaz de representar conhecimentos probabilsticos. Tambm proposta neste trabalho uma extenso das linguagens de comunicao atuais, que ser capaz de suportar as necessidades de comunicao de conhecimentos de natureza probabilsticas. So demonstradas as propriedades de compatibilidade do novo modelo lgico-probabilstico com o modelo puramente lgico atual, sendo demonstrado que teoremas vlidos no modelo atual continuam vlidos no novo modelo. O novo modelo definido como uma lgica probabilstica que estende a lgica modal dos modelos atuais. Para esta lgica probabilstica definido um sistema axiomtico e so demonstradas sua correo e completude. A completude demonstrada de forma relativa: se o sistema axiomtico da lgica modal original for completo, ento o sistema axiomtico da lgica probabilstica proposta como extenso tambm ser completo. A linguagem de comunicao proposta neste trabalho definida formalmente pela generalizao das teorias axiomticas de agncia e comunicao atuais para lidar com a comunicao de conhecimentos probabilsticos e pela definio de novos atos comunicativos especficos para este tipo de comunicao. Demonstra-se que esta linguagem compatvel com as linguagens atuais no caso no-probabilstico. Tambm definida uma nova linguagem para representao de contedos de atos de comunicao, baseada na lgica probabilstica usada como modelo semntico, que ser capaz de expressar conhecimentos probabilsticos e no probabilsticos de uma maneira uniforme. O grau de expressibilidade destas linguagens verificado por meio de duas aplicaes. Na primeira aplicao demonstra-se como a nova linguagem de contedos pode ser utilizada para representar conhecimentos probabilsticos expressos atravs da forma de representao de conhecimentos probabilsticos mais aceita atualmente, que so as Redes Bayesianas ou Redes de Crenas Probabilsticas. Na outra aplicao, so propostos protocolos de interao, baseados nos novos atos comunicativos, que so capazes de atender as necessidades de comunicao das operaes de consistncia de Redes Bayesianas secionadas (MSBNs, Multiple Sectioned Bayesian Networks) para o caso de sistemas multiagentes.
