Clustering techniques for patients suffering acquired brain injury inneuro personal trainer

The study of the effectiveness of the cognitive rehabilitation processes and the identification of cognitive profiles, in order to define comparable populations, is a controversial area, but concurrently it is strongly needed in order to improve therapies. There is limited evidence about cognitive rehabilitation efficacy. Many of the trials conclude that in spite of an apparent clinical good response, differences do not show statistical significance. The common feature in all these trials is heterogeneity among populations. In this situation, observational studies on very well controlled cohort of studies, together with innovative methods in knowledge extraction, could provide methodological insights for the design of more accurate comparative trials. Some correlation studies between neuropsychological tests and patients capacities have been carried out -1---2- and also correlation between tests and morphological changes in the brain -3-. The procedures efficacy depends on three main factors: the affectation profile, the scheduled tasks and the execution results. The relationship between them makes up the cognitive rehabilitation as a discipline, but its structure is not properly defined. In this work we present a clustering method used in Neuro Personal Trainer (NPT) to group patients into cognitive profiles using data mining techniques. The system uses these clusters to personalize treatments, using the patients assigned cluster to select which tasks are more suitable for its concrete needs, by comparing the results obtained in the past by patients with the same profile.

Neuronanatomy, neurology and Bayesian networks

Bayesian networks are data mining models with clear semantics and a sound theoretical foundation. In this keynote talk we will pinpoint a number of neuroscience problems that can be addressed using Bayesian networks. In neuroanatomy, we will show computer simulation models of dendritic trees and classification of neuron types, both based on morphological features. In neurology, we will present the search for genetic biomarkers in Alzheimer's disease and the prediction of health-related quality of life in Parkinson's disease. Most of these challenging problems posed by neuroscience involve new Bayesian network designs that can cope with multiple class variables, small sample sizes, or labels annotated by several experts.

Mejora de herramientas de Inteligencia de Negocio en su aplicación a la Gestión Educativa

Tras los distintos análisis diseñados por Jorge Beltrán Luna en el proyecto "Aplicación de Inteligencia de Negocio a la Gestión Educativa" [Beltrán2014] sobre el comportamiento de los alumnos de la Universidad Politécnica de Madrid en las asignaturas cursadas por estos durante el curso 2013-2014, se llegó a la conclusión que se debía desarrollar una aplicación web mediante la cual pudiesen configurarse estos análisis con distintos parámetros para adecuarlos a los requerimientos del usuario. Este proyecto ha cumplido con el objetivo anteriormente mencionado. Se ha desarrollado una aplicación web capaz de mostrar por medio de un navegador web, las gráficas y tablas generadas por el programa de minería de datos. Mediante esta aplicación el usuario puede realizar diversas funciones. Una de ellas es la de solicitar mediante el formulario recibido en la interfaz principal de la aplicación, la visualización de los resultados generados por el sistema de acuerdo con los parámetros seleccionados por el diseñador de los análisis. El usuario conseguirá observar los resultados que obtendría si ejecutase directamente los análisis desarrollados en el proyecto de Jorge Beltrán Luna [Beltrán2014] en la herramienta Rapidminer. Otra de las funciones que podría realizar el usuario sería la de realizar estos mismos análisis pero modificando sus parámetros de configuración para adecuar dichos análisis a los resultados que se quiere obtener. El resultado será el que se habría conseguido en la aplicación Rapidminer si se cambiasen los mismos parámetros que los modificados en la página web de este prototipo. Por último, se ha diseñado un botón con el cual el usuario podrá recuperar el último análisis realizado, con el fin de que no sea necesario esperar el tiempo que tarde en realizarse el análisis para visualizar los resultados. También se ha realizado una explicación detallada de la aplicación de la inteligencia de negocio en el ámbito educacional. ABSTRACT. After different analysis designed by Jorge Beltran Luna in the "Aplicación de Inteligencia de Negocio a la Gestión Educativa" [Beltrán2014] project on the behaviour of the students at the Universidad Politécnica de Madrid during the course 2013-14, the tutor of this project concluded that it should be interesting to develop a web application through which teachers could view and configure these analysis with different parameters. This project has fulfilled the aforementioned objective. A web application has been develop to show through a web browser, the graphs and charts generated by the data mining tool. Using this application, the user can perform various features. One of this features is to request, employing the formulary received in the main interface, to display an analysis according to the chosen parameters. The user will see the results that would be observed in case that the analysis had been directly executed using the project designed by Jorge Beltrán Luna [Beltrán2014] in the RapidMiner tool. Another feature that the user could perform would be to make these analysis modifying its settings Similar result would be obtained in the RapidMiner tool in the case that identical modifications were carried out in the configuration parameters. Finally, a button to allow with recall the last analysis has been implemented. It is not necessary to wait for the execution of this analysis to see newly the results. A detailed explanation on the usage of business intelligence in the educational field has also been performed.

Monitorización inteligente de redes de comunicación

El avance tecnológico de los últimos años ha aumentado la necesidad de guardar enormes cantidades de datos de forma masiva, llegando a una situación de desorden en el proceso de almacenamiento de datos, a su desactualización y a complicar su análisis. Esta situación causó un gran interés para las organizaciones en la búsqueda de un enfoque para obtener información relevante de estos grandes almacenes de datos. Surge así lo que se define como inteligencia de negocio, un conjunto de herramientas, procedimientos y estrategias para llevar a cabo la “extracción de conocimiento”, término con el que se refiere comúnmente a la extracción de información útil para la propia organización. Concretamente en este proyecto, se ha utilizado el enfoque Knowledge Discovery in Databases (KDD), que permite lograr la identificación de patrones y un manejo eficiente de las anomalías que puedan aparecer en una red de comunicaciones. Este enfoque comprende desde la selección de los datos primarios hasta su análisis final para la determinación de patrones. El núcleo de todo el enfoque KDD es la minería de datos, que contiene la tecnología necesaria para la identificación de los patrones mencionados y la extracción de conocimiento. Para ello, se utilizará la herramienta RapidMiner en su versión libre y gratuita, debido a que es más completa y de manejo más sencillo que otras herramientas como KNIME o WEKA. La gestión de una red engloba todo el proceso de despliegue y mantenimiento. Es en este procedimiento donde se recogen y monitorizan todas las anomalías ocasionadas en la red, las cuales pueden almacenarse en un repositorio. El objetivo de este proyecto es realizar un planteamiento teórico y varios experimentos que permitan identificar patrones en registros de anomalías de red. Se ha estudiado el repositorio de MAWI Lab, en el que se han almacenado anomalías diarias. Se trata de buscar indicios característicos anuales detectando patrones. Los diferentes experimentos y procedimientos de este estudio pretenden demostrar la utilidad de la inteligencia de negocio a la hora de extraer información a partir de un almacén de datos masivo, para su posterior análisis o futuros estudios. ABSTRACT. The technological progresses in the recent years required to store a big amount of information in repositories. This information is often in disorder, outdated and needs a complex analysis. This situation has caused a relevant interest in investigating methodologies to obtain important information from these huge data stores. Business intelligence was born as a set of tools, procedures and strategies to implement the "knowledge extraction". Specifically in this project, Knowledge Discovery in Databases (KDD) approach has been used. KDD is one of the most important processes of business intelligence to achieve the identification of patterns and the efficient management of the anomalies in a communications network. This approach includes all necessary stages from the selection of the raw data until the analysis to determine the patterns. The core process of the whole KDD approach is the Data Mining process, which analyzes the information needed to identify the patterns and to extract the knowledge. In this project we use the RapidMiner tool to carry out the Data Mining process, because this tool has more features and is easier to use than other tools like WEKA or KNIME. Network management includes the deployment, supervision and maintenance tasks. Network management process is where all anomalies are collected, monitored, and can be stored in a repository. The goal of this project is to construct a theoretical approach, to implement a prototype and to carry out several experiments that allow identifying patterns in some anomalies records. MAWI Lab repository has been selected to be studied, which contains daily anomalies. The different experiments show the utility of the business intelligence to extract information from big data warehouse.

Autonomous classification models in ubiquitous environments

Stream-mining approach is defined as a set of cutting-edge techniques designed to process streams of data in real time, in order to extract knowledge. In the particular case of classification, stream-mining has to adapt its behaviour to the volatile underlying data distributions, what has been called concept drift. Moreover, it is important to note that concept drift may lead to situations where predictive models become invalid and have therefore to be updated to represent the actual concepts that data poses. In this context, there is a specific type of concept drift, known as recurrent concept drift, where the concepts represented by data have already appeared in the past. In those cases the learning process could be saved or at least minimized by applying a previously trained model. This could be extremely useful in ubiquitous environments that are characterized by the existence of resource constrained devices. To deal with the aforementioned scenario, meta-models can be used in the process of enhancing the drift detection mechanisms used by data stream algorithms, by representing and predicting when the change will occur. There are some real-world situations where a concept reappears, as in the case of intrusion detection systems (IDS), where the same incidents or an adaptation of them usually reappear over time. In these environments the early prediction of drift by means of a better knowledge of past models can help to anticipate to the change, thus improving efficiency of the model regarding the training instances needed. By means of using meta-models as a recurrent drift detection mechanism, the ability to share concepts representations among different data mining processes is open. That kind of exchanges could improve the accuracy of the resultant local model as such model may benefit from patterns similar to the local concept that were observed in other scenarios, but not yet locally. This would also improve the efficiency of training instances used during the classification process, as long as the exchange of models would aid in the application of already trained recurrent models, that have been previously seen by any of the collaborative devices. Which it is to say that the scope of recurrence detection and representation is broaden. In fact the detection, representation and exchange of concept drift patterns would be extremely useful for the law enforcement activities fighting against cyber crime. Being the information exchange one of the main pillars of cooperation, national units would benefit from the experience and knowledge gained by third parties. Moreover, in the specific scope of critical infrastructures protection it is crucial to count with information exchange mechanisms, both from a strategical and technical scope. The exchange of concept drift detection schemes in cyber security environments would aid in the process of preventing, detecting and effectively responding to threads in cyber space. Furthermore, as a complement of meta-models, a mechanism to assess the similarity between classification models is also needed when dealing with recurrent concepts. In this context, when reusing a previously trained model a rough comparison between concepts is usually made, applying boolean logic. The introduction of fuzzy logic comparisons between models could lead to a better efficient reuse of previously seen concepts, by applying not just equal models, but also similar ones. This work faces the aforementioned open issues by means of: the MMPRec system, that integrates a meta-model mechanism and a fuzzy similarity function; a collaborative environment to share meta-models between different devices; a recurrent drift generator that allows to test the usefulness of recurrent drift systems, as it is the case of MMPRec. Moreover, this thesis presents an experimental validation of the proposed contributions using synthetic and real datasets.

Desarrollo de un sistema de arañas inteligentes para la extracción de datos de empresas

El presente proyecto de fin de grado es uno de los resultados generados en un proyecto de financiación privada por parte de Telefónica consistente en el desarrollo y posterior implantación de un sistema para minería de datos de empresas presentes en Internet. Este TFG surge a partir de un proyecto que el grupo de investigación AICU-LABS (Mercator) de la UPM ha desarrollado para Telefónica, y tiene como elemento principal el desarrollo de Agentes web (también llamados robots software, “softbots” o “crawlers”) capaces de obtener datos de empresas a partir de sus CIF a través de internet. El listado de empresas nos los proporciona Telefónica, y está compuesto por empresas que no son clientes de Telefónica en la actualidad. Nuestra misión es proporcionarles los datos necesarios (principalmente teléfono, correo electrónico y dirección de la empresa) para la creación de una base de datos de potenciales clientes. Para llevar a cabo esta tarea, se ha realizado una aplicación que, a partir de los CIF que nos proporcionan, busque información en internet y extraiga aquella que nos interese. Además se han desarrollado sistemas de validación de datos para ayudarnos a descartar datos no válidos y clasificar los datos según su calidad para así maximizar la calidad de los datos producidos por el robot. La búsqueda de datos se hará tanto en bases de datos online como, en caso de localizarlas, las propias páginas web de las empresas. ABSTRACT This Final Degree Project is one of the results obtained from a project funded by Telefónica. This project consists on the development and subsequent implantation of a system which performs data mining on companies operating on the Internet. This document arises from a project the research group AICU-LABS (Mercator) from the Universidad Politécnica de Madrid has developed for Telefónica. The main goal of this project is the creation of web agents (also known as “crawlers” or “web spiders”) able to obtain data from businesses through the Internet, knowing only their VAT identification number. The list of companies is given by Telefónica, and it is composed by companies that are not Telefónica’s customers today. Our mission is to provide the data required (mainly phone, email and address of the company) to create a database of potential customers. To perform this task, we’ve developed an application that, starting with the given VAT numbers, searches the web for information and extracts the data sought. In addition, we have developed data validation systems, that are capable of discarding low quality data and also sorting the data according to their quality, to maximize the quality of the results produced by the robot. We’ll use both the companies’ websites and external databases as our sources of information.

Development of a Dashboard for Sentiment Analysis of Football in Twitter based on Web Components and D3.js

This thesis is the result of a project whose objective has been to develop and deploy a dashboard for sentiment analysis of football in Twitter based on web components and D3.js. To do so, a visualisation server has been developed in order to present the data obtained from Twitter and analysed with Senpy. This visualisation server has been developed with Polymer web components and D3.js. Data mining has been done with a pipeline between Twitter, Senpy and ElasticSearch. Luigi have been used in this process because helps building complex pipelines of batch jobs, so it has analysed all tweets and stored them in ElasticSearch. To continue, D3.js has been used to create interactive widgets that make data easily accessible, this widgets will allow the user to interact with them and �filter the most interesting data for him. Polymer web components have been used to make this dashboard according to Google's material design and be able to show dynamic data in widgets. As a result, this project will allow an extensive analysis of the social network, pointing out the influence of players and teams and the emotions and sentiments that emerge in a lapse of time.

Case-based reasoning to support decision making for managing drinking water quality events in distribution systems

Peer reviewed

BIND—The Biomolecular Interaction Network Database

The Biomolecular Interaction Network Database (BIND; http://binddb.org) is a database designed to store full descriptions of interactions, molecular complexes and pathways. Development of the BIND 2.0 data model has led to the incorporation of virtually all components of molecular mechanisms including interactions between any two molecules composed of proteins, nucleic acids and small molecules. Chemical reactions, photochemical activation and conformational changes can also be described. Everything from small molecule biochemistry to signal transduction is abstracted in such a way that graph theory methods may be applied for data mining. The database can be used to study networks of interactions, to map pathways across taxonomic branches and to generate information for kinetic simulations. BIND anticipates the coming large influx of interaction information from high-throughput proteomics efforts including detailed information about post-translational modifications from mass spectrometry. Version 2.0 of the BIND data model is discussed as well as implementation, content and the open nature of the BIND project. The BIND data specification is available as ASN.1 and XML DTD.

The Comprehensive Microbial Resource

One challenge presented by large-scale genome sequencing efforts is effective display of uniform information to the scientific community. The Comprehensive Microbial Resource (CMR) contains robust annotation of all complete microbial genomes and allows for a wide variety of data retrievals. The bacterial information has been placed on the Web at http://www.tigr.org/CMR for retrieval using standard web browsing technology. Retrievals can be based on protein properties such as molecular weight or hydrophobicity, GC-content, functional role assignments and taxonomy. The CMR also has special web-based tools to allow data mining using pre-run homology searches, whole genome dot-plots, batch downloading and traversal across genomes using a variety of datatypes.

Changes in global gene expression patterns during development and maturation of the rat kidney

We set out to define patterns of gene expression during kidney organogenesis by using high-density DNA array technology. Expression analysis of 8,740 rat genes revealed five discrete patterns or groups of gene expression during nephrogenesis. Group 1 consisted of genes with very high expression in the early embryonic kidney, many with roles in protein translation and DNA replication. Group 2 consisted of genes that peaked in midembryogenesis and contained many transcripts specifying proteins of the extracellular matrix. Many additional transcripts allied with groups 1 and 2 had known or proposed roles in kidney development and included LIM1, POD1, GFRA1, WT1, BCL2, Homeobox protein A11, timeless, pleiotrophin, HGF, HNF3, BMP4, TGF-α, TGF-β2, IGF-II, met, FGF7, BMP4, and ganglioside-GD3. Group 3 consisted of transcripts that peaked in the neonatal period and contained a number of retrotransposon RNAs. Group 4 contained genes that steadily increased in relative expression levels throughout development, including many genes involved in energy metabolism and transport. Group 5 consisted of genes with relatively low levels of expression throughout embryogenesis but with markedly higher levels in the adult kidney; this group included a heterogeneous mix of transporters, detoxification enzymes, and oxidative stress genes. The data suggest that the embryonic kidney is committed to cellular proliferation and morphogenesis early on, followed sequentially by extracellular matrix deposition and acquisition of markers of terminal differentiation. The neonatal burst of retrotransposon mRNA was unexpected and may play a role in a stress response associated with birth. Custom analytical tools were developed including “The Equalizer” and “eBlot,” which contain improved methods for data normalization, significance testing, and data mining.

Análisis del modelo RFM según el método convencional y el método de las 2-Tuplas

En el siguiente Trabajo de Fin de Máster se pone en práctica la Minería de Datos (Data Mining), llevando a cabo una investigación de CRM (Customer Relationship Management) en la cual se analizan los comportamientos de compra de los clientes de una empresa que comercializa solo por internet (online). Este negocio es de origen español y mediante estos análisis podremos saber principalmente cuántos tipos de clientes posee y cómo son sus hábitos de compra para poder clasificarlos. Para ello, utilizaremos la segmentación RFM (Recency, Frequency, Monetary) que la calcularemos mediante dos metodologías muy importantes que son el Método Convencional y el Método de las 2-Tuplas. En el primer método realizaremos una clasificación numérica mediante quintiles que se numerarán de 1 a 5 tanto para la Recencia, la Frecuencia y el Valor Monetario, con los que podremos determinar el comportamiento de compra de cada cliente. En el segundo método veremos otra clasificación de los clientes más precisa, más detallada y con la ventaja que ofrece un valor lingüístico para poder entender mejor a que cluster pertenece cada cliente. Finalmente, realizaremos unos análisis de clusters con el método de “K-medias” con diferentes segmentos (entre 5 y 7 segmentos) que nos permitirán distinguir cuántos tipos de clientes tiene este negocio y cómo son con respecto a su hábito de compra. Todo esto con el fin de dar respuesta a este negocio sobre cómo es el comportamiento de compra de cada cliente, cuáles son los más importantes, cuáles son los menos importantes, cuántos han dejado de comprar, etc.

A model driven framework for geographic knowledge discovery

Geographic knowledge discovery (GKD) is the process of extracting information and knowledge from massive georeferenced databases. Usually the process is accomplished by two different systems, the Geographic Information Systems (GIS) and the data mining engines. However, the development of those systems is a complex task due to it does not follow a systematic, integrated and standard methodology. To overcome these pitfalls, in this paper, we propose a modeling framework that addresses the development of the different parts of a multilayer GKD process. The main advantages of our framework are that: (i) it reduces the design effort, (ii) it improves quality systems obtained, (iii) it is independent of platforms, (iv) it facilitates the use of data mining techniques on geo-referenced data, and finally, (v) it ameliorates the communication between different users.

El proceso de enseñanza-aprendizaje en la formación interuniversitaria: una experiencia a nivel de máster universitario

La incorporación del EEES provocó una infinidad de desafíos y retos a las Universidades que a día de hoy aún están siendo solucionados. Además, ha conllevado nuevas oportunidades para la formación de estudiantes pero también para las Universidades. Entre ellas, la formación interuniversitaria entre estados miembro de la UE. El EEES permite unificar a través del sistema ECTS la carga de trabajo de los estudiantes facilitando la propuesta de planes de estudios interuniversitarios. Sin embargo, surgen desafíos a la hora de llevarlos a la práctica. Independientemente de los retos en la propuesta de los planes de estudio, es necesario implementar procesos de enseñanza-aprendizaje que salven la distancia en el espacio físico entre el alumnado y el profesorado. En este artículo se presenta la experiencia docente de la asignatura e-home del Máster Machine Learning and Data Mining de la Universidad de Alicante y la Universidad Jean Monnet (Francia). En este caso, se combina la formación en aula presencial con formación en aula virtual a través de videoconferencia. La evaluación del método de enseñanza-aprendizaje propuesto utiliza la propia experiencia docente y encuestas realizadas a los alumnos para poner de manifiesto la ruptura de barreras espaciales y un éxito a nivel docente.

Estadística Aplicada a los Recursos Marinos. Clustering. Análisis de segmentación

Introducción al análisis con Clustering