José Francisco Aldana Montes

Full Automatic ANN Design: A Genetic Approach

ANN design is usually thought as a training problem to be solved for some predefined ANN structure and connectivity. Training methods arc very problem and ANN dependent. They are sometimes very accurate procedures but they work in narrow and restrictive domains. Thus the designer is faced to a wide diversity of multimodal and different training mechanisms. We have selected Genetic Algorithms as training procedures because of their robustness and their potential application to any ANN type training. Furthermore we have addressed the connectivity and structure definition problems in order to accomplish a full genetic ANN design. These three levels of design can work in parallel, thus achieving multilevel relationships to yield better ANNs. GRIAL is the tool used to test several new and known genetic techniques and operators. PARLOG is the Concurrent Logic Language used for the implementation in order to introduce new models for the genetic work and attain an intralevel distributed search as well as to parallelize any ANN evaluation.

URANO: An Object-Oriented Artificial Neural Network Simulation Tool

A highly flexible environment has been developed based on the object-oriented paradigm for modelling artificial neural networks (ANNs). This paper propose a hierarchy of classes that models ANN. The design of the hierarchy is characterized by a high degree of modularity, based on parametrizable data structures and autonomous modules. Composition rules of structures and methods enable to build, step by step, more complex structures from simple ones previously defined. One of the most relevant benefits obtained from using an object-oriented approach is related with the definition of multi-networks. URANO (Universe of ANN Object oriented) is a powerful software tool developed using the C++ programming language for building and using prototypes of ANNs in which inherited mechanism can be used for building new ANNs from already existing ones. It is easy to work with and its effectiveness and efficiency have been proven by applications.

Extending an Object Oriented Concurrent Logic Language for Neural Network Simulations

An extension of an object oriented concurrent logic language for a flexible and rapid prototyping of new neural network models and its simulation is proposed. The environment combines the logic and object oriented paradigms. On one hand the declarative expressiveness of the logic languages allows a natural representation of connectivity in neural models. On the other hand the object oriented approach helps us in our effort in obtaining higher level simulation tools, because the lower level characteristics of the language are hidden by this approach and any new model can be made from the currently existing ones. Furthermore, because our environment is based on a concurrent language, it is easy to specify the parallel characteristics of the neural network and to implement our tools in a parallel computer.

Design and architecture of the jMetaISP framework

jMetaISP is a framework for dynamic multi-objective Big Data optimization. It combines the jMetal multi-objective framework with the Apache Spark cluster computing system to allow the solving of dynamic optimization problems from a number of external streaming data sources in Big Data contexts. In this paper, we describe the current status of the jMetaISP project, focusing mainly in its design and internal architecture, with the aim of offering a comprehensive view of its main features to interested researchers. Among the covered features, we describe the main components of a jMetalSP application, including dynamic problems, dynamic algorithms, streaming data sources, and data consumers. For practical purposes, we describe two test cases to illustrate how to address dynamic combinatorial and dynamic continuous optimization problems by using the proposed framework.

DomESA: a novel approach for extending domain-oriented lexical relatedness calculations with domain-specific semantics

Being able to correctly model semantic relatedness between texts, and consequently the concepts represented by these texts, has become an important part of many intelligent information retrieval and knowledge processing systems. The need for such systems is especially evident within the biomedical domain, where the sheer amount of scientific publishing contributes to an information overflow. In this paper we present a novel method to approximate semantic relatedness in domain-focused settings. The approach is an extension to a well-known ESA (Explicit Semantic Analysis) method. Our extension successfully leverages the semantics of a domain-specific document corpus. We present the evaluation of the proposed method on a set of reference datasets, that are a de facto reference standard for the task of approximating biomedical semantic relatedness. The proposed method is evaluated in comparison with other state-of-the-art methods, as well as the baselines established with the original ESA method. The results of the experiments suggest that the proposed method combines the semantics of a general and domain-specific corpora to provide significant improvements over the original method.

MORPHY: A Multiobjective Software Tool for Phylogenetic Inference of Protein Coded Sequences

Most of software solutions for phylogenetic inference try to find the best phylogenetic tree according to one reconstruction criterion, maximum parsimony or maximum likelihood, making the exploration of different hypothesis based on these two features a complex process. In this work, we present a novel software tool for phylogenetic inference based on a multiobjective approach called MORPHY, which searches for a set of compromise solutions according to the criteria of maximum parsimony and maximum likelihood at the same time. This tool not only works with DNA sequences, but also allows to deal with protein coded datasets. It is implemented using the multiobjective and phylogenetic features of the software MO-Phylogenetics, and the program outputs are a set of optimized trees in Newick format. A consensus tree from all the obtained solutions can also be produced. MORPHY’s executable, source code, and sample datasets are publicly available at the web repository: https://github.com/KhaosResearch/MORPHY.

A Repository of Semantic Open EHR Archetypes

This paper describes a repository of openEHR archetypes that have been translated to OWL. In the work presented here, five different CKMs (Clinical Knowledge Managers) have been downloaded and the archetypes have been translated to OWL. This translation is based on an existing translator that has been improved to solve programming problems with certain structures. As part of the repository a tool has been developed to keep it always up-to-date. So, any change in one of the CKMs (addition, elimination or even change of an archetype) will involve translating the changed archetypes once more. The repository is accessible through a Web interface (http://www.openehr.es/).