Juan Llorens

Decision support system for forest fire protection in the Euro-Mediterranean region

This paper describes the development of a decision support system (DSS) for prevention planning and emergency management of forest fire events that incorporates weather data management, a geographical data viewer, a priori danger forecasting and fire propagation modeling, automatic fire detection, and optimal resource dispatching. Collection, input, storage, management, and analysis of the information rely on advanced and automated methodologies using remote sensing, GPS, digital mapping, and geographic information systems. The results included short-term dynamic fire danger indices developed for improved and realistic prevention and pre-suppression planning. An automatic fire detection technology based on infrared video was developed and successfully tested on site. Several models for understanding fire propagation on forest fires have been proposed for practical application. Additionally, a DSS was developed with the innovation of covering wildland fire hazard management entirely, providing a complete coverage of technical and administrative activities that support decision makers in real time. The DSS was tested for high fire seasons in two different sites in South Europe.

Melodic similarity through shape similarity

We present a new geometric model to compute the melodic similarity of symbolic musical pieces. Melodies are represented as splines in the pitch-time plane, and their similarity is computed as the similarity of their shape. The model is very intuitive and it is transposition and time scale invariant. We have implemented it with a local alignment algorithm over sequences of n-grams that define spline spans. An evaluation with the MIREX 2005 collections shows that the model performs very well, obtaining the best effectiveness scores ever reported for these collections. Three systems based on this new model were evaluated in MIREX 2010, and the three systems obtained the best results.

A methodology for the classification of quality of requirements using machine learning techniques

ContextOne of the most important factors in the development of a software project is the quality of their requirements. Erroneous requirements, if not detected early, may cause many serious problems, such as substantial additional costs, failure to meet the expected objectives and delays in delivery dates. For these reasons, great effort must be devoted in requirements engineering to ensure that the projects requirements results are of high quality. One of the aims of this discipline is the automatic processing of requirements for assessing their quality; this aim, however, results in a complex task because the quality of requirements depends mostly on the interpretation of experts and the necessities and demands of the project at hand. ObjectiveThe objective of this paper is to assess the quality of requirements automatically, emulating the assessment that a quality expert of a project would assess. MethodThe proposed methodology is based on the idea of learning based on standard metrics that represent the characteristics that an expert takes into consideration when deciding on the good or bad quality of requirements. Using machine learning techniques, a classifier is trained with requirements earlier classified by the expert, which then is used for classifying newly provided requirements. ResultsWe present two approaches to represent the methodology with two situations of the problem in function of the requirement corpus learning balancing, obtaining different results in the accuracy and the efficiency in order to evaluate both representations. The paper demonstrates the reliability of the methodology by presenting a case study with requirements provided by the Requirements Working Group of the INCOSE organization. ConclusionsA methodology that evaluates the quality of requirements written in natural language is presented in order to emulate the quality that the expert would provide for new requirements, with 86.1 of average in the accuracy.

Ontology-Assisted Systems Engineering Process with Focus in the Requirements Engineering Process

Problems found in the current Systems Engineering with focus in the Requirements Engineering Process shown that it could be improved using ontologies for aiding in the process. Requirements engineering in the Systems Engineering process is enhanced and quality of requirements enriched as well, improving Systems Engineering capabilities clearly can result in better Project Performance. One of that is the Requirement improvement and of course the benefit goes to the whole process of development. The more correct, complete and consistent it is, the best performance it will have and ontologies enable a more exhaustive and fast quality process.

Tagging for improved semantic interpretation of XML

The development of the Semantic Web depends on agreed and unambiguous knowledge representations, on the availability and accessibility of knowledge, and on retrieval capabilities. The scarce agreement on knowledge representation and the lack of techniques to process semantic structures in web search engines makes it impossible to perform contextualized conceptual retrieval. These limitations imply that users must know beforehand the existence and location of this knowledge to be able to retrieve it. Thus, different ad-hoc knowledge representations and metadata vocabularies, scarcely formalized and agreed on, have been published, hindering the reuse and interoperability. Our proposal has as its main goal the elaboration of different views to improve the representation of semantic documents. This approach facilitates the management and retrieval of heterogeneous semantic schemas by means of a multilevel ontological structure and the alignment with a reference ontology that provides conceptual retrieval and reuse of knowledge.

Training Initiative for New Software/Enterprise Architects: An Ontological Approach

In this paper, we describe the importance of new software/enterprise architects in the discipline of software architecture and enterprise architecture. Both are often idealized as super heroes with a lot of qualities that are very infrequent in contemporary people. The enterprise/software architect role could be assumed by a group of people able to manage the qualities for the role. In any case, even a group or a single person must be educated in the discipline by training courses, new methodologies of learning, or traditional university studies. In order to improve the process of becoming a new architect we propose a methodology based on ontological structures and reinforcement learning.

Syntactic-Semantic Extraction of Patterns Applied to the US and European Patents Domain.

Nowadays, there are many scientific inventions referring to any topic like medicine, technology, economics, finance, banking, computer science, and so on. These inventions are suggested as patents to the agencies working in US and Europe for the registration and revision of the patent applications. But, the job of reviewing the patents might be complicated because every day the quantity of it is bigger and bigger. And also, the amount of work dedicated writing a proper application might be intricate and needs several revisions from investor and examiners. This revision job might have costs for the inventor because they don’t know the proper language for writing the application in the formal mode used. As part of a solution, one approach to minimize the impact of this fact and increase the success of the reviewing process is aid the human reviewer and also inventors with a set of patterns created using Natural Language Processing techniques that accelerate the review just looking in the massive set of registration any similar one already patented and on the other hand aid the inventor writing in the formal manner the application.

Automatic Pattern Generator of Natural Language Text Applied in Public Health

At the moment, a huge amount of scientific articles is available, referring to a wide variety of topics like medicine, technology, economics, finance, and so on. Scientific papers show results of scientific interest and also present the evaluation and interpretation of relevant arguments. Due to the fact that these papers are created with a high frequency it is feasible to analyze how people write in a given domain. Within the discipline of natural language processing there are different approaches to analyze large amounts of text corpus. Identification patterns with semantic elements in a text, let us classify and examine the corpus to facilitate interpretation and management of information through computers. At the moment, a semiautomatic or automatic way to generate natural language patterns is not available or quite complicated. In the paper, it is shown how a tool developed for this research is tested in a domain of public health. The results obtained – by means of a tool and aided by graphs – provide groups of words that are used (to determine if they come from a specific vocabulary), most common grammatical categories, most repeated words in a domain, patterns found, and frequency of patterns found. A domain of public health has been selected containing 800 papers concerning different topics referring to genetics. The topics include mutations, genetic deafness, DNA, trinucleotide, suppressor genes, among others. An ontology of public health has been used to provide the basis of the study.

Natural Language Processing System Applied in Public Health for Assessment of an Automatic Analysis of Patterns Generator

Nowadays, there are many scientific articles referring to any topic like medicine, technology, economics, finance, and so on. These articles are better known as papers, they represent the evaluation and interpretation of different arguments, showing results of scientific interest. At the end, most of these are published in magazines, books, journals, etc. Due to the fact that these papers are created with a higher frequency it is feasible to analyse how people write in the same domain. At the level of structure and with the help of graphs some of the results that can be found are: groups of words that are used (to determine if they come from a specific vocabulary), most common grammatical categories, most repeated words in a domain, patterns found, and frequency of patterns found. This research has been created to fulfil these needs. A domain of public health has been selected and it is composed of 800 papers about different topics referring to genetics such as mutations, genetic deafness, DNA, trinucleotide, suppressor genes, among others; and an ontology of public health has been used to provide the basis of the study.

How to Build Ontologies for Requirements Systems Engineering Projects Aiding the Quality Management Process

Knowledge is centric to systems engineering, the knowledge management process must take into account that a Systems Knowledge Repository (SKR) exists as a key element for either quality improvement, traceability support and, in summary, reuse purposes. Requirements engineering in the Systems Engineering process is enhanced by using knowledge systems and quality of requirements enriched as well. The more correct, complete and consistent a requirement is, the best performance it will have and knowledge systems enable a more exhaustive and fast quality process. A knowledge management process is proposed and it is guided by a requirements domain based example using a Knowledge Management tool supporting the whole process.