Automatic learning for the classification of chemical reactions and in statistical thermodynamics

This Thesis describes the application of automatic learning methods for a) the classification of organic and metabolic reactions, and b) the mapping of Potential Energy Surfaces(PES). The classification of reactions was approached with two distinct methodologies: a representation of chemical reactions based on NMR data, and a representation of chemical reactions from the reaction equation based on the physico-chemical and topological features of chemical bonds. NMR-based classification of photochemical and enzymatic reactions. Photochemical and metabolic reactions were classified by Kohonen Self-Organizing Maps (Kohonen SOMs) and Random Forests (RFs) taking as input the difference between the 1H NMR spectra of the products and the reactants. The development of such a representation can be applied in automatic analysis of changes in the 1H NMR spectrum of a mixture and their interpretation in terms of the chemical reactions taking place. Examples of possible applications are the monitoring of reaction processes, evaluation of the stability of chemicals, or even the interpretation of metabonomic data. A Kohonen SOM trained with a data set of metabolic reactions catalysed by transferases was able to correctly classify 75% of an independent test set in terms of the EC number subclass. Random Forests improved the correct predictions to 79%. With photochemical reactions classified into 7 groups, an independent test set was classified with 86-93% accuracy. The data set of photochemical reactions was also used to simulate mixtures with two reactions occurring simultaneously. Kohonen SOMs and Feed-Forward Neural Networks (FFNNs) were trained to classify the reactions occurring in a mixture based on the 1H NMR spectra of the products and reactants. Kohonen SOMs allowed the correct assignment of 53-63% of the mixtures (in a test set). Counter-Propagation Neural Networks (CPNNs) gave origin to similar results. The use of supervised learning techniques allowed an improvement in the results. They were improved to 77% of correct assignments when an ensemble of ten FFNNs were used and to 80% when Random Forests were used. This study was performed with NMR data simulated from the molecular structure by the SPINUS program. In the design of one test set, simulated data was combined with experimental data. The results support the proposal of linking databases of chemical reactions to experimental or simulated NMR data for automatic classification of reactions and mixtures of reactions. Genome-scale classification of enzymatic reactions from their reaction equation. The MOLMAP descriptor relies on a Kohonen SOM that defines types of bonds on the basis of their physico-chemical and topological properties. The MOLMAP descriptor of a molecule represents the types of bonds available in that molecule. The MOLMAP descriptor of a reaction is defined as the difference between the MOLMAPs of the products and the reactants, and numerically encodes the pattern of bonds that are broken, changed, and made during a chemical reaction. The automatic perception of chemical similarities between metabolic reactions is required for a variety of applications ranging from the computer validation of classification systems, genome-scale reconstruction (or comparison) of metabolic pathways, to the classification of enzymatic mechanisms. Catalytic functions of proteins are generally described by the EC numbers that are simultaneously employed as identifiers of reactions, enzymes, and enzyme genes, thus linking metabolic and genomic information. Different methods should be available to automatically compare metabolic reactions and for the automatic assignment of EC numbers to reactions still not officially classified. In this study, the genome-scale data set of enzymatic reactions available in the KEGG database was encoded by the MOLMAP descriptors, and was submitted to Kohonen SOMs to compare the resulting map with the official EC number classification, to explore the possibility of predicting EC numbers from the reaction equation, and to assess the internal consistency of the EC classification at the class level. A general agreement with the EC classification was observed, i.e. a relationship between the similarity of MOLMAPs and the similarity of EC numbers. At the same time, MOLMAPs were able to discriminate between EC sub-subclasses. EC numbers could be assigned at the class, subclass, and sub-subclass levels with accuracies up to 92%, 80%, and 70% for independent test sets. The correspondence between chemical similarity of metabolic reactions and their MOLMAP descriptors was applied to the identification of a number of reactions mapped into the same neuron but belonging to different EC classes, which demonstrated the ability of the MOLMAP/SOM approach to verify the internal consistency of classifications in databases of metabolic reactions. RFs were also used to assign the four levels of the EC hierarchy from the reaction equation. EC numbers were correctly assigned in 95%, 90%, 85% and 86% of the cases (for independent test sets) at the class, subclass, sub-subclass and full EC number level,respectively. Experiments for the classification of reactions from the main reactants and products were performed with RFs - EC numbers were assigned at the class, subclass and sub-subclass level with accuracies of 78%, 74% and 63%, respectively. In the course of the experiments with metabolic reactions we suggested that the MOLMAP / SOM concept could be extended to the representation of other levels of metabolic information such as metabolic pathways. Following the MOLMAP idea, the pattern of neurons activated by the reactions of a metabolic pathway is a representation of the reactions involved in that pathway - a descriptor of the metabolic pathway. This reasoning enabled the comparison of different pathways, the automatic classification of pathways, and a classification of organisms based on their biochemical machinery. The three levels of classification (from bonds to metabolic pathways) allowed to map and perceive chemical similarities between metabolic pathways even for pathways of different types of metabolism and pathways that do not share similarities in terms of EC numbers. Mapping of PES by neural networks (NNs). In a first series of experiments, ensembles of Feed-Forward NNs (EnsFFNNs) and Associative Neural Networks (ASNNs) were trained to reproduce PES represented by the Lennard-Jones (LJ) analytical potential function. The accuracy of the method was assessed by comparing the results of molecular dynamics simulations (thermal, structural, and dynamic properties) obtained from the NNs-PES and from the LJ function. The results indicated that for LJ-type potentials, NNs can be trained to generate accurate PES to be used in molecular simulations. EnsFFNNs and ASNNs gave better results than single FFNNs. A remarkable ability of the NNs models to interpolate between distant curves and accurately reproduce potentials to be used in molecular simulations is shown. The purpose of the first study was to systematically analyse the accuracy of different NNs. Our main motivation, however, is reflected in the next study: the mapping of multidimensional PES by NNs to simulate, by Molecular Dynamics or Monte Carlo, the adsorption and self-assembly of solvated organic molecules on noble-metal electrodes. Indeed, for such complex and heterogeneous systems the development of suitable analytical functions that fit quantum mechanical interaction energies is a non-trivial or even impossible task. The data consisted of energy values, from Density Functional Theory (DFT) calculations, at different distances, for several molecular orientations and three electrode adsorption sites. The results indicate that NNs require a data set large enough to cover well the diversity of possible interaction sites, distances, and orientations. NNs trained with such data sets can perform equally well or even better than analytical functions. Therefore, they can be used in molecular simulations, particularly for the ethanol/Au (111) interface which is the case studied in the present Thesis. Once properly trained, the networks are able to produce, as output, any required number of energy points for accurate interpolations.

TRAILER project overview: tagging, recognition and acknowledgment of informal learning experiences

The evolution of new technology and its increasing use, have for some years been making the existence of informal learning more and more transparent, especially among young and older adults in both Higher Education and workplace contexts. However, the nature of formal and non-formal, course-based, approaches to learning has made it hard to accommodate these informal processes satisfactorily, and although technology bring us near to the solution, it has not yet achieved. TRAILER project aims to address this problem by developing a tool for the management of competences and skills acquired through informal learning experiences, both from the perspective of the user and the institution or company. This paper describes the research and development main lines of this project.

Modelling text file evaluation processes

Text file evaluation is an emergent topic in e-learning that responds to the shortcomings of the assessment based on questions with predefined answers. Questions with predefined answers are formalized in languages such as IMS Question & Test Interoperability Specification (QTI) and supported by many e-learning systems. Complex evaluation domains justify the development of specialized evaluators that participate in several business processes. The goal of this paper is to formalize the concept of a text file evaluation in the scope of the E-Framework – a service oriented framework for development of e-learning systems maintained by a community of practice. The contribution includes an abstract service type and a service usage model. The former describes the generic capabilities of a text file evaluation service. The later is a business process involving a set of services such as repositories of learning objects and learning management systems.

Modelling a network of heterogeneous e-learning systems

In recent years emerged several initiatives promoted by educational organizations to adapt Service Oriented Architectures (SOA) to e-learning. These initiatives commonly named eLearning Frameworks share a common goal: to create flexible learning environments by integrating heterogeneous systems already available in many educational institutions. However, these frameworks were designed for integration of systems participating in business like processes rather than on complex pedagogical processes as those related to automatic evaluation. Consequently, their knowledge bases lack some fundamental components that are needed to model pedagogical processes. The objective of the research described in this paper is to study the applicability of eLearning frameworks for modelling a network of heterogeneous eLearning systems, using the automatic evaluation of programming exercises as a case study. The paper surveys the existing eLearning frameworks to justify the selection of the e-Framework. This framework is described in detail and identified the necessary components missing from its knowledge base, more precisely, a service genre, expression and usage model for an evaluation service. The extensibility of the framework is tested with the definition of this service. A concrete model for evaluation of programming exercises is presented as a validation of the proposed approach.

Brownian bridge and other path-dependent Gaussian processes vectorial simulation

The iterative simulation of the Brownian bridge is well known. In this article, we present a vectorial simulation alternative based on Gaussian processes for machine learning regression that is suitable for interpreted programming languages implementations. We extend the vectorial simulation of path-dependent trajectories to other Gaussian processes, namely, sequences of Brownian bridges, geometric Brownian motion, fractional Brownian motion, and Ornstein-Ulenbeck mean reversion process.

Interoperability frameworks for competencies and learning outcomes: databases vs ontologies

The changes introduced into the European Higher Education Area (EHEA) by the Bologna Process, together with renewed pedagogical and methodological practices, have created a new teaching-learning paradigm: Student-Centred Learning. In addition, the last few years have been characterized by the application of Information Technologies, especially the Semantic Web, not only to the teaching-learning process, but also to administrative processes within learning institutions. On one hand, the aim of this study was to present a model for identifying and classifying Competencies and Learning Outcomes and, on the other hand, the computer applications of the information management model were developed, namely a relational Database and an Ontology.

Comportamento do consumidor digital: o caso do e-learning nas licenciaturas do ensino superior português

Dissertação de Mestrado apresentado ao Instituto de Contabilidade e Administração do Porto para a obtenção do grau de Mestre em Marketing Digital, sob orientação do professor Doutor Manuel Silva

Machine learning Gaussian short rate

Dissertação para obtenção do Grau de Doutor em Estatística e Gestão do Risco

Learning Activities with Semantic. Hypermedia in Higher Education

The increasing use of information and communication technologies (ICT) in diverse professional and personal contexts calls for new knowledge, and a set of abilities, competences and attitudes, for an active and participative citizenship. In this context it is acknowledged that universities have an important role innovating in the educational use of digital media to promote an inclusive digital literacy. The educational potential of digital technologies and resources has been recognized by both researchers and practitioners. Multiple pedagogical models and research approaches have already contributed to put in evidence the importance of adapting instructional and learning practices and processes to concrete contexts and educational goals. Still, academic and scientific communities believe further investments in ICT research is needed in higher education. This study focuses on educational models that may contribute to support digital technology uses, where these can have cognitive and educational relevance when compared to analogical technologies. A teaching and learning model, centered in the active role of the students in the exploration, production, presentation and discussion of interactive multimedia materials, was developed and applied using the internet and exploring emergent semantic hypermedia formats. The research approach focused on the definition of design principles for developing class activities that were applied in three different iterations in undergraduate courses from two institutions, namely the University of Texas at Austin, USA and the University of Lisbon, Portugal. The analysis of this study made possible to evaluate the potential and efficacy of the model proposed and the authoring tool chosen in the support of metacognitive skills and attitudes related to information structuring and management, storytelling and communication, using computers and the internet.

Gamification e game-based learning: uma abordagem lúdica à aprendizagem

Relatório de estágio de mestrado em Ensino de Informática

Discrimination of Brazilian propolis according to the seasoning using chemometrics and machine learning based on UV-Vis scanning data

Propolis is a chemically complex biomass produced by honeybees (Apis mellifera) from plant resins added of salivary enzymes, beeswax, and pollen. The biological activities described for propolis were also identified for donor plants resin, but a big challenge for the standardization of the chemical composition and biological effects of propolis remains on a better understanding of the influence of seasonality on the chemical constituents of that raw material. Since propolis quality depends, among other variables, on the local flora which is strongly influenced by (a)biotic factors over the seasons, to unravel the harvest season effect on the propolis chemical profile is an issue of recognized importance. For that, fast, cheap, and robust analytical techniques seem to be the best choice for large scale quality control processes in the most demanding markets, e.g., human health applications. For that, UV-Visible (UV-Vis) scanning spectrophotometry of hydroalcoholic extracts (HE) of seventy-three propolis samples, collected over the seasons in 2014 (summer, spring, autumn, and winter) and 2015 (summer and autumn) in Southern Brazil was adopted. Further machine learning and chemometrics techniques were applied to the UV-Vis dataset aiming to gain insights as to the seasonality effect on the claimed chemical heterogeneity of propolis samples determined by changes in the flora of the geographic region under study. Descriptive and classification models were built following a chemometric approach, i.e. principal component analysis (PCA) and hierarchical clustering analysis (HCA) supported by scripts written in the R language. The UV-Vis profiles associated with chemometric analysis allowed identifying a typical pattern in propolis samples collected in the summer. Importantly, the discrimination based on PCA could be improved by using the dataset of the fingerprint region of phenolic compounds ( = 280-400m), suggesting that besides the biological activities of those secondary metabolites, they also play a relevant role for the discrimination and classification of that complex matrix through bioinformatics tools. Finally, a series of machine learning approaches, e.g., partial least square-discriminant analysis (PLS-DA), k-Nearest Neighbors (kNN), and Decision Trees showed to be complementary to PCA and HCA, allowing to obtain relevant information as to the sample discrimination.

Aprendizaje durante la ontogenia temprana: Estudio de la participación del sistema opioide en la modulación de aprendizajes mediados por el etanol. LEARNING AND EARLY ONTOGENY: ANALYSIS OF THE INVOLVEMENT OF OPIOID SYSTEM IN FETAL AND INFANTILE ETHANOL-MEDIATED EXPERIENCIES.

Este plan de trabajos es continuidad de una línea de investigación centrada en evaluar los mecanismos responsables de la adquisición, expresión y persistencia de experiencias con el etanol. A partir de ello, indagar acerca de los efectos de esta experiencia sobre comportamientos de búsqueda y autoadministración de etanol en neonatos e infantes de rata. Se pretende analizar la participación del sistema opiáceo en los mecanismos implicados en una memoria fetal y/o infantil, generada como consecuencia de la exposición etílica. En una primera etapa, nos proponemos establecer de qué manera experiencias prenatales con la droga modulan el patrón de auto-administración de alcohol y otros reforzadores, como sacarosa. En este primer bloque de experimentos realizaremos manipulaciones fetales para determinar con mayor grado de especificidad la posible acción del sistema opiáceo en los mecanismos de adquisición de una memoria etílica prenatal. Se realizarán administraciones de etanol y el antagonista opiáceo, directamente a nivel fetal, y se evaluará esta experiencia en un paradigma de condicionamiento neonatal positivo, mediado por la droga. De acuerdo a la evidencia previa, esperamos que la exposición prenatal con la droga facilite la expresión de conductas de consumo y búsqueda del etanol o hacia las claves que señalizan al psicotrópico, tanto durante la infancia como en el neonato. A su vez, cuando la droga es presentada bajo los efectos de un antagonista opiáceo esperamos que estas conductas muestren un perfil similar a las desplegadas por sujetos controles. El segundo bloque de experimentos ha sido ideado con el objeto de indagar acerca de la posible participación del sistema opiáceo en la modulación de los aspectos reforzantes de la droga, a través de un esquema de auto-administración etílica infantil. Se utilizará un paradigma de condicionamiento instrumental adaptado para ratas infantes que consta de dos instancias, una de adquisición de la conducta instrumental (DPs 14-17) en la cual los animales reciben un pulso de refuerzo, como consecuencia de la ejecución de la conducta operante. En una segunda fase se analiza el patrón de búsqueda del reforzador ya que se registra la respuesta instrumental, sin que ocurra el refuerzo por la misma. Para analizar la participación del sistema opiáceo, durante la fase de adquisición de la conducta operante (DPs 16 y 17) los animales serán re-expuestos a mínimas cantidades del reforzador, bajo los efectos de un antagonista opiáceo, momentos previos al ensayo instrumental correspondiente para cada uno de estos días (Exp. 3). Esperamos que el bloqueo del sistema opiáceo, durante esta re-exposición al etanol, sea suficiente para disminuir el patrón de respuesta instrumental hacia el refuerzo etílico. Un último experimento incorporará un tercer evento de re-exposición al etanol -bajo los efectos del antagonista- previo al ensayo de extinción de la conducta instrumental (DP 18). Este nuevo evento tiene por objeto analizar la participación de este sistema neurobiológico en los mecanismos de búsqueda de etanol. Si el sistema opiáceo participa en la modulación de patrones tanto de búsqueda como consumatorios del reforzamiento por etanol, se espera que la re-exposición a la droga bajo los efectos del antagonista, inhiba estas respuestas tanto durante la sesión de adquisición, como de extinción de la conducta operante. Este proyecto intenta profundizar en el conocimiento de los mecanismos que regulan reconocimiento, aceptación, búsqueda y consumo de etanol, como consecuencia de experiencias tempranas con la droga. A su vez, es importante identificar y estudiar los sistemas neurobiológicos involucrados en estos mecanismos. Es por ello que se intenta determinar el rol que ejerce el sistema opiáceo en la adquisición de estas experiencias etílicas a nivel fetal e infantil, que se conoce promueven la búsqueda y el consumo de la droga. Our work is directed to analyze the involvement of the opioid system in the generation of pre- and early postnatal ethanol-related memories. As a first step, maternal manipulations with ethanol will be done. Infants will be evaluated in a paradigm of infantile self-administration of different reinforcers (ethanol, sucrose or water), employing a model of operant conditioning adapted to infant rats. A second experiment will be conducted in order to analyze if a central administration of ethanol, directly to the fetus, modifies subsequent patterns of neonatal conditioned responses to an artificial nipple, mediated by ethanol reinforcing effects. Fetal presentation of ethanol will be accompanied with the injection of an opioid antagonist in order to analyze the involvement of this system in acquisition processes of a fetal ethanol-mediated memory. A second set of studies will be conducted to analyze appetitive and consummatory behaviors in an infant model of ethanol self-administration. Involvement of opioid system in the acquisition or expression of this experience will be also inquired. Infant rats (PDs14-17) have to display a target behavior (nose-poke) to gain access to 5% sucrose or 3.75% ethanol. On PD18 an extinction session will be included. At PDs16-17, 6-hr before training, pups will be re-exposed to ethanol under opioid antagonism effects (naloxone). In a follow up experiment, a re-exposure trial will be included at PD18. Prior extinction, pups will receive naloxone and will be re-exposed to ethanol. We aim to observe if opioid system is modulating etha¬nol reinforcing effects, in terms of both appetitive and consummatory behaviors.

Learning the ropes and embracing the rules: CFSP institutions as arenas for learning and strategic socialisation

The aim of this article is to analyse those situations in which learning and socialisation take place within the context of the Common Foreign and Security Policy (CFSP), in particular, at the level of experts in the Council Working Groups. Learning can explain the institutional development of CFSP and changes in the foreign policies of the Member States. Some scope conditions for learning and channels of institutionalisation are identified. Socialisation, resulting from learning within a group, is perceived as a strategic action by reflective actors. National diplomats, once they arrive in Brussels, learn the new code of conduct of their Working Groups. They are embedded in two environments and faced with two logics: the European one in the Council and the national one in the Ministries of Foreign Affairs (MFA). The empirical evidence supports the argument that neither rational nor sociological approaches alone can account for these processes.

Autonomous underwater vehicle control using reinforcement learning policy search methods

Autonomous underwater vehicles (AUV) represent a challenging control problem with complex, noisy, dynamics. Nowadays, not only the continuous scientific advances in underwater robotics but the increasing number of subsea missions and its complexity ask for an automatization of submarine processes. This paper proposes a high-level control system for solving the action selection problem of an autonomous robot. The system is characterized by the use of reinforcement learning direct policy search methods (RLDPS) for learning the internal state/action mapping of some behaviors. We demonstrate its feasibility with simulated experiments using the model of our underwater robot URIS in a target following task

Adaptive intelligent agent approach to guide the web navigation on the PLAN-G distance learning platform

Hypermedia systems based on the Web for open distance education are becoming increasinglypopular as tools for user-driven access learning information. Adaptive hypermedia is a new direction in research within the area of user-adaptive systems, to increase its functionality by making it personalized [Eklu 961. This paper sketches a general agents architecture to include navigationaladaptability and user-friendly processes which would guide and accompany the student during hislher learning on the PLAN-G hypermedia system (New Generation Telematics Platform to Support Open and Distance Learning), with the aid of computer networks and specifically WWW technology [Marz 98-1] [Marz 98-2]. The PLAN-G actual prototype is successfully used with some informatics courses (the current version has no agents yet). The propased multi-agent system, contains two different types of adaptive autonomous software agents: Personal Digital Agents {Interface), to interacl directly with the student when necessary; and Information Agents (Intermediaries), to filtrate and discover information to learn and to adapt navigation space to a specific student