Design and Validation of a Light Inference System to Support Embedded Context Reasoning

Embedded context management in resource-constrained devices (e.g. mobile phones, autonomous sensors or smart objects) imposes special requirements in terms of lightness for data modelling and reasoning. In this paper, we explore the state-of-the-art on data representation and reasoning tools for embedded mobile reasoning and propose a light inference system (LIS) aiming at simplifying embedded inference processes offering a set of functionalities to avoid redundancy in context management operations. The system is part of a service-oriented mobile software framework, conceived to facilitate the creation of context-aware applications?it decouples sensor data acquisition and context processing from the application logic. LIS, composed of several modules, encapsulates existing lightweight tools for ontology data management and rule-based reasoning, and it is ready to run on Java-enabled handheld devices. Data management and reasoning processes are designed to handle a general ontology that enables communication among framework components. Both the applications running on top of the framework and the framework components themselves can configure the rule and query sets in order to retrieve the information they need from LIS. In order to test LIS features in a real application scenario, an ?Activity Monitor? has been designed and implemented: a personal health-persuasive application that provides feedback on the user?s lifestyle, combining data from physical and virtual sensors. In this case of use, LIS is used to timely evaluate the user?s activity level, to decide on the convenience of triggering notifications and to determine the best interface or channel to deliver these context-aware alerts.

Reducing communication overhead for cooperative localization using nonparametric belief propagation

A number of methods for cooperative localization has been proposed, but most of them provide only location estimate, without associated uncertainty. On the other hand, nonparametric belief propagation (NBP), which provides approximated posterior distributions of the location estimates, is expensive mostly because of the transmission of the particles. In this paper, we propose a novel approach to reduce communication overhead for cooperative positioning using NBP. It is based on: i) communication of the beliefs (instead of the messages), ii) approximation of the belief with Gaussian mixture of very few components, and iii) censoring. According to our simulations results, these modifications reduce significantly communication overhead while providing the estimates almost as accurate as the transmission of the particles.

Permutation tests for nonparametric detection

In this paper, the authors provide a methodology to design nonparametric permutation tests and, in particular, nonparametric rank tests for applications in detection. In the first part of the paper, the authors develop the optimization theory of both permutation and rank tests in the Neyman?Pearson sense; in the second part of the paper, they carry out a comparative performance analysis of the permutation and rank tests (detectors) against the parametric ones in radar applications. First, a brief review of some contributions on nonparametric tests is realized. Then, the optimum permutation and rank tests are derived. Finally, a performance analysis is realized by Monte-Carlo simulations for the corresponding detectors, and the results are shown in curves of detection probability versus signal-to-noise ratio

Adaptive neuro-fuzzy inference systems (ANFIS) for supervised classification of competence in MIS based on psychomotor skills

INTRODUCTION: Objective assessment of motor skills has become an important challenge in minimally invasive surgery (MIS) training.Currently, there is no gold standard defining and determining the residents' surgical competence.To aid in the decision process, we analyze the validity of a supervised classifier to determine the degree of MIS competence based on assessment of psychomotor skills METHODOLOGY: The ANFIS is trained to classify performance in a box trainer peg transfer task performed by two groups (expert/non expert). There were 42 participants included in the study: the non-expert group consisted of 16 medical students and 8 residents (< 10 MIS procedures performed), whereas the expert group consisted of 14 residents (> 10 MIS procedures performed) and 4 experienced surgeons. Instrument movements were captured by means of the Endoscopic Video Analysis (EVA) tracking system. Nine motion analysis parameters (MAPs) were analyzed, including time, path length, depth, average speed, average acceleration, economy of area, economy of volume, idle time and motion smoothness. Data reduction was performed by means of principal component analysis, and then used to train the ANFIS net. Performance was measured by leave one out cross validation. RESULTS: The ANFIS presented an accuracy of 80.95%, where 13 experts and 21 non-experts were correctly classified. Total root mean square error was 0.88, while the area under the classifiers' ROC curve (AUC) was measured at 0.81. DISCUSSION: We have shown the usefulness of ANFIS for classification of MIS competence in a simple box trainer exercise. The main advantage of using ANFIS resides in its continuous output, which allows fine discrimination of surgical competence. There are, however, challenges that must be taken into account when considering use of ANFIS (e.g. training time, architecture modeling). Despite this, we have shown discriminative power of ANFIS for a low-difficulty box trainer task, regardless of the individual significances between MAPs. Future studies are required to confirm the findings, inclusion of new tasks, conditions and sample population.

Inference Models in DNA Computing

La computación molecular es una disciplina que se ocupa del diseño e implementación de dispositivos para el procesamiento de información sobre un sustrato biológico, como el ácido desoxirribonucleico (ADN), el ácido ribonucleico (ARN) o las proteínas. Desde que Watson y Crick descubrieron en los años cincuenta la estructura molecular del ADN en forma de doble hélice, se desencadenaron otros descubrimientos, como las enzimas de restricción o la reacción en cadena de la polimerasa (PCR), contribuyendo de manera determinante a la irrupción de la tecnología del ADN recombinante. Gracias a esta tecnología y al descenso vertiginoso de los precios de secuenciación y síntesis del ADN, la computación biomolecular pudo abandonar su concepción puramente teórica. El trabajo presentado por Adleman (1994) logró resolver un problema de computación NP-completo (El Problema del Camino de Hamilton dirigido) utilizando únicamente moléculas de ADN. La gran capacidad de procesamiento en paralelo ofrecida por las técnicas del ADN recombinante permitió a Adleman ser capaz de resolver dicho problema en tiempo polinómico, aunque a costa de un consumo exponencial de moléculas de ADN. Utilizando algoritmos de fuerza bruta similares al utilizado por Adleman se logró resolver otros problemas NP-completos, como por ejemplo el de Satisfacibilidad de Fórmulas Lógicas / SAT (Lipton, 1995). Pronto se comprendió que la computación biomolecular no podía competir en velocidad ni precisión con los ordenadores de silicio, por lo que su enfoque y objetivos se centraron en la resolución de problemas con aplicación biomédica (Simmel, 2007), dejando de lado la resolución de problemas clásicos de computación. Desde entonces se han propuesto diversos modelos de dispositivos biomoleculares que, de forma autónoma (sin necesidad de un bio-ingeniero realizando operaciones de laboratorio), son capaces de procesar como entrada un sustrato biológico y proporcionar una salida también en formato biológico: procesadores que aprovechan la extensión de la polimerasa (Hagiya et al., 1997), autómatas que funcionan con enzimas de restricción (Benenson et al., 2001) o con deoxiribozimas (Stojanovic et al., 2002), o circuitos de hibridación competitiva (Yurke et al., 2000). Esta tesis presenta un conjunto de modelos de dispositivos de ácidos nucleicos capaces de implementar diversas operaciones de computación lógica aprovechando técnicas de computación biomolecular (hibridación competitiva del ADN y reacciones enzimáticas) con aplicaciones en diagnóstico genético. El primer conjunto de modelos, presentados en el Capítulo 5 y publicados en Sainz de Murieta and Rodríguez-Patón (2012b), Rodríguez-Patón et al. (2010a) y Sainz de Murieta and Rodríguez-Patón (2010), define un tipo de biosensor que usa hebras simples de ADN para codificar reglas sencillas, como por ejemplo "SI hebra-ADN-1 Y hebra-ADN-2 presentes, ENTONCES enfermedad-B". Estas reglas interactúan con señales de entrada (ADN o ARN de cualquier tipo) para producir una señal de salida (también en forma de ácido nucleico). Dicha señal de salida representa un diagnóstico, que puede medirse mediante partículas fluorescentes técnicas FRET) o incluso ser un tratamiento administrado en respuesta a un conjunto de síntomas. El modelo presentado en el Capítulo 5, publicado en Rodríguez-Patón et al. (2011), es capaz de ejecutar cadenas de resolución sobre fórmulas lógicas en forma normal conjuntiva. Cada cláusula de una fórmula se codifica en una molécula de ADN. Cada proposición p se codifica asignándole una hebra simple de ADN, y la correspondiente hebra complementaria a la proposición ¬p. Las cláusulas se codifican incluyendo distintas proposiciones en la misma hebra de ADN. El modelo permite ejecutar programas lógicos de cláusulas Horn aplicando múltiples iteraciones de resolución en cascada, con el fin de implementar la función de un nanodispositivo autónomo programable. Esta técnica también puede emplearse para resolver SAP sin ayuda externa. El modelo presentado en el Capítulo 6 se ha publicado en publicado en Sainz de Murieta and Rodríguez-Patón (2012c), y el modelo presentado en el Capítulo 7 se ha publicado en (Sainz de Murieta and Rodríguez-Patón, 2013c). Aunque explotan métodos de computación biomolecular diferentes (hibridación competitiva de ADN en el Capítulo 6 frente a reacciones enzimáticas en el 7), ambos modelos son capaces de realizar inferencia Bayesiana. Funcionan tomando hebras simples de ADN como entrada, representando la presencia o la ausencia de un indicador molecular concreto (una evidencia). La probabilidad a priori de una enfermedad, así como la probabilidad condicionada de una señal (o síntoma) dada la enfermedad representan la base de conocimiento, y se codifican combinando distintas moléculas de ADN y sus concentraciones relativas. Cuando las moléculas de entrada interaccionan con las de la base de conocimiento, se liberan dos clases de hebras de ADN, cuya proporción relativa representa la aplicación del teorema de Bayes: la probabilidad condicionada de la enfermedad dada la señal (o síntoma). Todos estos dispositivos pueden verse como elementos básicos que, combinados modularmente, permiten la implementación de sistemas in vitro a partir de sensores de ADN, capaces de percibir y procesar señales biológicas. Este tipo de autómatas tienen en la actualidad una gran potencial, además de una gran repercusión científica. Un perfecto ejemplo fue la publicación de (Xie et al., 2011) en Science, presentando un autómata biomolecular de diagnóstico capaz de activar selectivamente el proceso de apoptosis en células cancerígenas sin afectar a células sanas.

Cooperative localization in mobile networks using nonparametric variants of belief propagation

Of the many state-of-the-art methods for cooperative localization in wireless sensor networks (WSN), only very few adapt well to mobile networks. The main problems of the well-known algorithms, based on nonparametric belief propagation (NBP), are the high communication cost and inefficient sampling techniques. Moreover, they either do not use smoothing or just apply it o ine. Therefore, in this article, we propose more flexible and effcient variants of NBP for cooperative localization in mobile networks. In particular, we provide: i) an optional 1-lag smoothing done almost in real-time, ii) a novel low-cost communication protocol based on package approximation and censoring, iii) higher robustness of the standard mixture importance sampling (MIS) technique, and iv) a higher amount of information in the importance densities by using the population Monte Carlo (PMC) approach, or an auxiliary variable. Through extensive simulations, we confirmed that all the proposed techniques outperform the standard NBP method.

Nonparametric generalized belief propagation based on pseudo-junction tree for cooperative localization in wireless networks

Non-parametric belief propagation (NBP) is a well-known message passing method for cooperative localization in wireless networks. However, due to the over-counting problem in the networks with loops, NBP’s convergence is not guaranteed, and its estimates are typically less accurate. One solution for this problem is non-parametric generalized belief propagation based on junction tree. However, this method is intractable in large-scale networks due to the high-complexity of the junction tree formation, and the high-dimensionality of the particles. Therefore, in this article, we propose the non-parametric generalized belief propagation based on pseudo-junction tree (NGBP-PJT). The main difference comparing with the standard method is the formation of pseudo-junction tree, which represents the approximated junction tree based on thin graph. In addition, in order to decrease the number of high-dimensional particles, we use more informative importance density function, and reduce the dimensionality of the messages. As by-product, we also propose NBP based on thin graph (NBP-TG), a cheaper variant of NBP, which runs on the same graph as NGBP-PJT. According to our simulation and experimental results, NGBP-PJT method outperforms NBP and NBP-TG in terms of accuracy, computational, and communication cost in reasonably sized networks.

Adaptable Bayesian classifier for spatiotemporal nonparametric moving object detection strategies

Electronic devices endowed with camera platforms require new and powerful machine vision applications, which commonly include moving object detection strategies. To obtain high-quality results, the most recent strategies estimate nonparametrically background and foreground models and combine them by means of a Bayesian classifier. However, typical classifiers are limited by the use of constant prior values and they do not allow the inclusion of additional spatiodependent prior information. In this Letter, we propose an alternative Bayesian classifier that, unlike those reported before, allows the use of additional prior information obtained from any source and depending on the spatial position of each pixel.

PV power forecast using a nonparametric PV model

Forecasting the AC power output of a PV plant accurately is important both for plant owners and electric system operators. Two main categories of PV modeling are available: the parametric and the nonparametric. In this paper, a methodology using a nonparametric PV model is proposed, using as inputs several forecasts of meteorological variables from a Numerical Weather Forecast model, and actual AC power measurements of PV plants. The methodology was built upon the R environment and uses Quantile Regression Forests as machine learning tool to forecast AC power with a confidence interval. Real data from five PV plants was used to validate the methodology, and results show that daily production is predicted with an absolute cvMBE lower than 1.3%.

Region-based moving object detection using spatially conditioned nonparametric models in a GPU

A novel GPU-based nonparametric moving object detection strategy for computer vision tools requiring real-time processing is proposed. An alternative and efficient Bayesian classifier to combine nonparametric background and foreground models allows increasing correct detections while avoiding false detections. Additionally, an efficient region of interest analysis significantly reduces the computational cost of the detections.

Comparison of parametric and nonparametric methods to map oligogenes by linkage

A sample of 95 sib pairs affected with insulin-dependent diabetes and typed with their normal parents for 28 markers on chromosome 6 has been analyzed by several methods. When appropriate parameters are efficiently estimated, a parametric model is equivalent to the β model, which is superior to nonparametric alternatives both in single point tests (as found previously) and in multipoint tests. Theory is given for meta-analysis combined with allelic association, and problems that may be associated with errors of map location and/or marker typing are identified. Reducing by multipoint analysis the number of association tests in a dense map can give a 3-fold reduction in the critical lod, and therefore in the cost of positional cloning.

Ecological inference

Ecological inference is the process of drawing conclusions about individual-level behavior from aggregate-level data. Recent advances involve the combination of statistical and deterministic means to produce such inferences.

What We Don't Know Can Hurt Us: Clinical Inference Bias and Constructed Memories of Sexual Abuse.

The false memory/recovered memory debate, research regarding the malleability of memory, and the current lack of methods for validating recovered memories all support the view that heightened care is required of therapists dealing with clients whom they suspect have been sexually abused. The judgmental heuristics that underlie the major clinical inference biases of confirmatory bias, biased covariation, base rate fallacies, and schematic processing errors are all relevant to the processes leading to therapist-client constructions of memories of sexual abuse. Suggestions for minimizing each of these biases are offered. Personal motivations of the client and client suggestibility are factors that may contribute to the construction of memories of sexual abuse, and suggestions for minimizing the impact of these motivations are offered. In conclusion, general suggestions for minimizing the impact of clinical inference biases within the sexual abuse treatment context are summarized.

Parametric & nonparametric approaches to event studies: an application to a hotel’s market value

The main objective of this paper is twofold: on the one hand, to analyse the impact that the announcement of the opening of a new hotel has on the performance of its chain by carrying out an event study, and on the other hand, to compare the results of two different approaches to this method: a parametric specification based on the autoregressive conditional heteroskedasticity models to estimate the market model, and a nonparametric approach, which implies employing Theil’s nonparametric regression technique, which in turn, leads to the so-called complete nonparametric approach to event studies. The results that the empirical application arrives at are noteworthy as, on average, the reaction to such news releases is highly positive, both approaches reaching the same level of significance. However, a word of caution must be said when one is not only interested in detecting whether the market reacts, but also in obtaining an exhaustive calculation of the abnormal returns to further examine its determining factors.

Inference of binding affinity from neuronal receptors in humans

Tese de mestrado, Bioinformática e Biologia Computacional (Bioinformática), Universidade de Lisboa, Faculdade de Ciências, 2016