Search

Cognitive accessibility aims to make content more accessible for people with intellectual and learning disabilities. By following existing standards, it is possible to improve cognitive accessibility, making text content more readable and understandable while, at the same time, designing a more accessible user interface. In this work, cognitive accessibility design patterns and their application in the Easier web system's user interface are presented. The Easier web system, which is targeted at people with cognitive disabilities, provides a tool that assists in understanding text content. In addition to the monitoring of design techniques, user tests have been carried out in order to provide a preliminary evaluation of the level of cognitive accessibility of the Easier system's interface.

Background: In the Big Data era there is an increasing need to fully exploit and analyse the huge quantity of information available about health. Natural Language Processing (NLP) technologies can contribute to extract relevant information from unstructured data contained in Electronic Health Records (EHR) such as clinical notes, patient’s discharge summaries and radiology reports among others. Extracted information could help in health-related decision making processes. Named entity recognition (NER) devoted to detect important concepts in texts (diseases, symptoms, drugs, etc.) is a crucial task in information extraction processes especially in languages other than English. In this work, we develop a deep learning-based NLP pipeline for biomedical entity extraction in Spanish clinical narrative. Methods: We explore the use of contextualized word embeddings to enhance named entity recognition in Spanish language clinical text, particularly of pharmacological substances, compounds, and proteins. Various combinations of word and sense embeddings were tested on the evaluation corpus of the PharmacoNER 2019 task, the Spanish Clinical Case Corpus (SPACCC). This data set consists of clinical case sections derived from open access Spanish-language medical publications. Results: NER system integrates in-domain pre-trained Flair and FastText word embeddings, byte-pairwise encoded and the bi-LSTM-based character word embeddings. The system yielded the best performance measure with F-score of 90.84%. Error analysis showed that the main source of errors for the best model is the newly detected false positive entities with the half of that amount of errors belonged to longer than the actual ones detected entities. Conclusions: Our study shows that our deep-learning-based system with domain-specific contextualized embeddings coupled with stacking of complementary embeddings yields superior performance over the system with integrated standard and general-domain word embeddings. With this system, we achieve performance competitive with the state-of-the-art.

People with intellectual, language and learning disabilities face accessibility barriers when reading texts with complex words. Following accessibility guidelines, complex words can be identified, and easy synonyms and definitions can be provided for them as reading aids. To offer support to these reading aids, a lexical simplification system for Spanish has been developed and is presented in this article. The system covers the complex word identification (CWI) task and offers replacement candidates with the substitute generation and selection (SG/SS) task. These tasks have followed machine learning techniques and contextual embeddings using Easy Reading and Plain Language resources, such as dictionaries and corpora. Additionally, due to the polysemy present in the language, the system provides definitions for complex words, which are disambiguated by a rule-based method supported by a state-of-the-art embedding resource. This system is integrated into a web system that provides an easy way to improve the readability and comprehension of Spanish texts. The results obtained are satisfactory; in the CWI task, better results were obtained than with other systems that used the same dataset. The SG/SS task results are comparable to similar works in the English language and provide a solid starting point to improve this task for the Spanish language. Finally, the results of the disambiguation process evaluation were good when evaluated by a linguistic expert. These findings represent an additional advancement in the lexical simplification of texts in Spanish and in a generic domain using easy-to-read resources, among others, to provide systematic support to compliance with accessibility guidelines This work was supported in part by the Madrid Government (Comunidad de Madrid-Spain) under the Multiannual Agreement with UC3M in the line of Excellence of University Professors under Grant EPUC3M17, in part by the context of the V PRICIT (Regional Programme of Research and Technological Innovation), and in part by the Accessible Technologies Award-INDRA Technologies and the Fundación Universia (www.tecnologiasaccesibles.com)

Thanks to Palestine Technical University-Kadoorie and Deep EMR project(TIN2017-87548-C2-1-R)for partially funding this work.

People with cognitive, language and learning disabilities face accessibility barriers when reading texts with complex or specialized words. In order for these needs to be addressed, and in accordance with accessibility guidelines, it is beneficial to provide the definitions of complex words to the user. In this sense, human language can, at times, be ambiguous, and many words may be interpreted in multiple ways depending on the context. To offer a correct definition, it is often necessary to carry out Word Sense Disambiguation. In this paper, a system that is based on Natural Language Processing and Language Resources in the field of easy reading to provide a contextualized definition for a complex word is presented. An expert linguistic, specialized in easy reading and plain language, has evaluated the Word Sense Disambiguation system. This research work is part of the EASIER project that offers an accessible platform to provide users with the easiest possible Spanish text to understand and read.

Difficulties in understanding texts that contain unusual words can create accessibility barriers for people with cognitive, language and learning disabilities. In this work, the EASIER system, a web system that provides various tools which improve cognitive accessibility, is presented. From a text in Spanish, complex words are detected and synonyms, a definition and a pictogram are offered for each complex word detected. Language and accessibility resources were used, such as easy-to-read dictionaries. The web system can be accessed from both desktop computers and mobile devices. Moreover, a browser extension is also offered.

As the world moves towards an information society, it is not possible to consider the concept of e-Government that is restricted to the ability of just a few. At present, it is common to have to fill out forms electronically in order to carry out administrative procedures. The web interfaces of public transactional e-Services must be accessible by system users, regardless of whether they are disabled or not, according to the regulatory framework. This accessibility is not always taken into consideration in online forms in a process flow. One possible reason for this is the fact that guides are not available to assist in the design and evaluation of their accessibility. It is necessary for the transactional processes on Government e-Service websites to be accessible and well designed, in accordance with the current standards. In order to achieve this goal, the standard Web Content Accessibility Guidelines (WCAG) must be complied with, according to legislation. However, these guidelines do not cover many of the obstacles faced by people with disabilities, as well as the general public, while filling out web forms, thus leaving accessibility problems undetected and unsolved. For this reason, web accessibility standards are not enough and should be supplemented by official government design patterns focused on transactional e-Services and online forms. In this article, we propose a support resource, which includes the accessibility criteria to be followed while designing and evaluating government transactional e-Service websites that incorporate online forms in the process flow. For the purpose of evaluating the reliability of the method and explaining it, a study was carried out on a sample of Spanish government e-Services used for making appointments in order to detect accessibility and design problems; this is presented as an example of good practices.

Background Drug Package Leaflets (DPLs) provide information for patients on how to safely use medicines. Pharmaceutical companies are responsible for producing these documents. However, several studies have shown that patients usually have problems in understanding sections describing posology (dosage quantity and prescription), contraindications and adverse drug reactions. An ultimate goal of this work is to provide an automatic approach that helps these companies to write drug package leaflets in an easy-to-understand language. Natural language processing has become a powerful tool for improving patient care and advancing medicine because it leads to automatically process the large amount of unstructured information needed for patient care. However, to the best of our knowledge, no research has been done on the automatic simplification of drug package leaflets. In a previous work, we proposed to use domain terminological resources for gathering a set of synonyms for a given target term. A potential drawback of this approach is that it depends heavily on the existence of dictionaries, however these are not always available for any domain and language or if they exist, their coverage is very scarce. To overcome this limitation, we propose the use of word embeddings to identify the simplest synonym for a given term. Word embedding models represent each word in a corpus with a vector in a semantic space. Our approach is based on assumption that synonyms should have close vectors because they occur in similar contexts. Results In our evaluation, we used the corpus EasyDPL (Easy Drug Package Leaflets), a collection of 306 leaflets written in Spanish and manually annotated with 1400 adverse drug effects and their simplest synonyms. We focus on leaflets written in Spanish because it is the second most widely spoken language on the world, but as for the existence of terminological resources, the Spanish language is usually less prolific than the English language. Our experiments show an accuracy of 38.5% using word embeddings. Conclusions This work provides a promising approach to simplify DPLs without using terminological resources or parallel corpora. Moreover, it could be easily adapted to different domains and languages. However, more research efforts are needed to improve our approach based on word embedding because it does not overcome our previous work using dictionaries yet.

Interactive multimodal information retrieval systems (IMIR) increase the capabilities of traditional search systems, by adding the ability to retrieve information of different types (modes) and from different sources. This article describes a formal model for interactive multimodal information retrieval. This model includes formal and widespread definitions of each component of an IMIR system. A use case that focuses on information retrieval regarding sports validates the model, by developing a prototype that implements a subset of the features of the model. Adaptive techniques applied to the retrieval functionality of IMIR systems have been defined by analysing past interactions using decision trees, neural networks, and clustering techniques. This model includes a strategy for selecting sources and combining the results obtained from every source. After modifying the strategy of the prototype for selecting sources, the system is re-evaluated using classification techniques. This evaluation compares the normalised discounted cumulative gain (NDCG) measure obtained using two different approaches: the multimodal system using a baseline strategy based on predefined rules as a source selection strategy, and the same multimodal system with the functionality adapted by past user interactions. In the adapted system, a final value of 81,54% was obtained for the NDCG.

Artificial neural networks are highly flexible and efficient tools in the approximation of time series patterns. In recent years, more than 5,000 studies oriented to the use of neural networks in time series forecasting have been evidenced in the extant literature. However, the methodology used for its specification and construction still involves a lot of trial and error or is inherited from econometric and statistical procedures that do not fit perfectly to the characteristics of the time series. This is especially true when they present non-linear behavior; moreover, it is not designed for working with neural networks. The objective of this chapter is to present a five-step guide for the specification, design, and validation of a neural network model for forecasting time series.

Companies must deal with a high uncertainty caused by the characteristics of the markets and the economic, political, and social environment in which they offer their products and services. These characteristics are defined by the preferences of the consumers, which have a high variety coupled with the digital era. On the other hand, there is the necessity to implement measures that align the companies with the sustainability concepts, because of both legislations as well as the image that the customer could have of them. Due to this context, the organizations must find a way to optimize process and structures that require high flexibility given the need of combining perfect innovation, customization, standardization, and sustainability. Part of this planning process is the construction of forecast models that allows predicting with high precisión. In this chapter, a theoretical exposition is done and a literature revision of machine learning techniques is applied to try to solve the forecasting problem with special emphasis in neural networks and Case-Based Reasoning - CBR.

Today, Individuals Can Access An Ever-Increasing Number Of Services Via The Internet. However, Only When All People Are Able To Completely Access The Internet Can A Digital Society Be Considered Universal. We Present A Proposal To Harmonize Accessibility Standards That All Countries Must Adhere To. This work is supported by the DeepEMR project (TIN2017-87548-C2-1-R) and the Spanish Subtitling and Audio Description Center.

Individuals with intellectual disabilities face accessibility barriers due to the fact that text contents found on web pages are often not accessible. In order to prevent this situation from occurring, accessibility guidelines as Easy to Read, WCAG 2.1 and Plain Language guidelines must be followed. This paper presents a study of accessibility guidelines and a system which performs lexical simplification in order to systematize compliance with accessibility guidelines, as well as generating an accessible user interface with the simplified text.

This work presents a two-stage deep learning system for Named Entity Recognition (NER) and Relation Extraction (RE) from medical texts. These tasks are a crucial step to many natural language understanding applications in the biomedical domain. Automatic medical coding of electronic medical records, automated summarizing of patient records, automatic cohort identification for clinical studies, text simplification of health documents for patients, early detection of adverse drug reactions or automatic identification of risk factors are only a few examples of the many possible opportunities that the text analysis can offer in the clinical domain. In this work, our efforts are primarily directed towards the improvement of the pharmacovigilance process by the automatic detection of drug-drug interactions (DDI) from texts. Moreover, we deal with the semantic analysis of texts containing health information for patients. Our two-stage approach is based on Deep Learning architectures. Concretely, NER is performed combining a bidirectional Long Short-Term Memory (Bi-LSTM) and a Conditional Random Field (CRF), while RE applies a Convolutional Neural Network (CNN). Since our approach uses very few language resources, only the pre-trained word embeddings, and does not exploit any domain resources (such as dictionaries or ontologies), this can be easily expandable to support other languages and clinical applications that require the exploitation of semantic information (concepts and relationships) from texts. During the last years, the task of DDI extraction has received great attention by the BioNLP community. However, the problem has been traditionally evaluated as two separate subtasks: drug name recognition and extraction of DDIs. To the best of our knowledge, this is the first work that provides an evaluation of the whole pipeline. Moreover, our system obtains state-of-the-art results on the eHealth-KD challenge, which was part of the Workshop on Semantic Analysis at SEPLN (TASS-2018).

Natural Language Processing-based systems to monitor social media streams.MeaningCloud text analytics service.Integration of drug and adverse effects semantic resources.Drug-adverse effects relation extraction.Data visualization in time lines and aggregated data.Named entities recognition and relation extraction evaluation. In the last years, the habit of discussing healthcare issues with family and friends, even with unknown people, in the context of social networks has increased and processing user generated content has become a new challenge. This can help in on-line crowd surveillance for different applications (pharmacovigilance and filtering health contents in blogs among others) as well as extracting knowledge from unstructured text sources. In this article, a system that monitors health social media streams is described. It is based on several text analytics processes supported, among others, by MeaningCloud, a commercial platform which provides meaning extraction from texts in a Software as a Service mode. In this architecture, several domain resources are integrated to detect drugs and drug effects such as CIMA (official information about authorized drugs in Spain maintained by the Spanish Agency of Medicines and Health Products), MedDRA (Medical Dictionary for Regulatory Activities) and the SpanishDrugEffectDB database that contains relations between drugs and effects. Different ways of visualizing data considering time lines and aggregated data have been implemented. In order to show performance, an evaluation has been carried out over Named Entities Recognition (NER) and Relation Extraction (RE) tasks.

Background Negation and speculation are critical elements in natural language processing (NLP)-related tasks, such as information extraction, as these phenomena change the truth value of a proposition. In the clinical narrative that is informal, these linguistic facts are used extensively with the objective of indicating hypotheses, impressions, or negative findings. Previous state-of-the-art approaches addressed negation and speculation detection tasks using rule-based methods, but in the last few years, models based on machine learning and deep learning exploiting morphological, syntactic, and semantic features represented as spare and dense vectors have emerged. However, although such methods of named entity recognition (NER) employ a broad set of features, they are limited to existing pretrained models for a specific domain or language. Objective As a fundamental subsystem of any information extraction pipeline, a system for cross-lingual and domain-independent negation and speculation detection was introduced with special focus on the biomedical scientific literature and clinical narrative. In this work, detection of negation and speculation was considered as a sequence-labeling task where cues and the scopes of both phenomena are recognized as a sequence of nested labels recognized in a single step. Methods We proposed the following two approaches for negation and speculation detection: (1) bidirectional long short-term memory (Bi-LSTM) and conditional random field using character, word, and sense embeddings to deal with the extraction of semantic, syntactic, and contextual patterns and (2) bidirectional encoder representations for transformers (BERT) with fine tuning for NER. Results The approach was evaluated for English and Spanish languages on biomedical and review text, particularly with the BioScope corpus, IULA corpus, and SFU Spanish Review corpus, with F-measures of 86.6%, 85.0%, and 88.1%, respectively, for NeuroNER and 86.4%, 80.8%, and 91.7%, respectively, for BERT. Conclusions These results show that these architectures perform considerably better than the previous rule-based and conventional machine learning–based systems. Moreover, our analysis results show that pretrained word embedding and particularly contextualized embedding for biomedical corpora help to understand complexities inherent to biomedical text.

In this work, we introduce a Deep Learning architecture for pharmaceutical and chemical Named Entity Recognition in Spanish clinical cases texts. We propose a hybrid model approach based on two Bidirectional Long Short-Term Memory (Bi-LSTM) network and Conditional Random Field (CRF) network using character, word, concept and sense embeddings to deal with the extraction of semantic, syntactic and morphological features. The approach was evaluated on the PharmaCoNER Corpus obtaining an F-measure of 85.24% for subtask 1 and 49.36% for subtask2. These results prove that deep learning methods with specific domain embedding representations can outperform the state-of-the-art approaches.

This work was supported by TrendMiner project [FP7-ICT287863] and by eGovernAbility-Access project (TIN2014-52665-C2-2-R).

Natural Language Processing (NLP) techniques can provide an interesting way to mine the growing biomedical literature, and a promising approach for new knowledge discovery. However, the major bottleneck in this area is that these systems rely on specific resources providing the domain knowledge. Domain ontologies provide a contextual framework and a semantic representation of the domain, and they can contribute to a better performance of current NLP systems. However, their contribution to information extraction has not been well studied yet. The aim of this paper is to provide insights into the potential role that domain ontologies can play in NLP. To do this, the authors apply the drug-drug interactions ontology (DINTO) to named entity recognition and relation extraction from pharmacological texts. The authors use the DDI corpus, a gold-standard for the development and evaluation of IE systems in this domain, and evaluate their results in the framework of the last SemEval-2013 DDI Extraction task.

Nowadays, consumption of video content is extremely high and the corresponding technology must be accessible in order for people with disabilities to be able to access it. For this reason, the user agents or media players used to access video content must be accessible. This poster presents a checklist that includes indicators which can assist in the design and evaluation of accessible media players.

[EN] 'Communication' between abiotic nanoscale chemical systems is an almost-unexplored field with enormous potential. Here we show the design and preparation of a chemical communication system based on enzyme-powered Janus nanoparticles, which mimics an interactive model of communication. Cargo delivery from one nanoparticle is governed by the biunivocal communication with another nanoparticle, which involves two enzymatic processes and the interchange of chemical messengers. The conceptual idea of establishing communication between nanodevices opens the opportunity to develop complex nanoscale systems capable of sharing information and cooperating., A. L.-L. is grateful to 'La Caixa' Banking Foundation for his PhD fellowship. We wish to thank the Spanish Government (MINECO Projects MAT2015-64139-C4-1, CTQ2014-58989-P and CTQ2015-71936-REDT and AGL2015-70235-C2-2-R) and the Generalitat Valenciana (Project PROMETEOII/2014/047) for support. The Comunidad de Madrid (S2013/MIT-3029, Programme NANOAVANSENS) is also gratefully acknowledged.

Display Omitted DDIExtraction aims to the extraction of drug-drug interactions from texts.Non-linear kernel-based methods overcome linear SVMs.Most results are statistically significant.Resolution of complex sentences could lead to better performance. The DDIExtraction Shared Task 2013 is the second edition of the DDIExtraction Shared Task series, a community-wide effort to promote the implementation and comparative assessment of natural language processing (NLP) techniques in the field of the pharmacovigilance domain, in particular, to address the extraction of drug-drug interactions (DDI) from biomedical texts. This edition has been the first attempt to compare the performance of Information Extraction (IE) techniques specific for each of the basic steps of the DDI extraction pipeline. To attain this aim, two main tasks were proposed: the recognition and classification of pharmacological substances and the detection and classification of drug-drug interactions. DDIExtraction 2013 was held from January to June 2013 and attracted wide attention with a total of 14 teams (6 of the teams participated in the drug name recognition task, while 8 participated in the DDI extraction task) from 7 different countries. For the task of the recognition and classification of pharmacological names, the best system achieved an F1 of 71.5%, while, for the detection and classification of DDIs, the best result was an F1 of 65.1%. The results show advances in the state of the art and demonstrate that significant challenges remain to be resolved. This paper focuses on the second task (extraction of DDIs) and examines its main challenges, which have yet to be resolved.

The increase in rich and interactive content on the Web, such as video and audio, has brought about a growth in the number and type of users that access it. This means that this content should be accessible to anyone, including people with disabilities. Therefore, it should be accompanied by media alternatives (captions, audio description, sign language, extended audio description, etc.), which should be handled by an accessible user agent that provides support for them. This paper presents: (1) an approach with accessibility requirements following standards to include accessibility in a user agent for delivering accessible multimedia content based on it, (2) a conceptualization of these requirements that provides an abstract user interface which may be integrated into the development process, after that, (3) a check of the integration of the accessibility requirements in the models, and (4) finally, some lessons learned and conclusions are shown.

This works is at: 5th International Conference on Corpus Linguistics (CILC2013), took place 2013, March 14-16, in Alicante (Spain). Conference web site: http://web.ua.es/en/cilc2013/ Natural language processing of pharmacological texts includes recognition of drug names and extraction of relationships between them. To this purpose, pharmacological annotated corpora are required. These corpora are usually semantically annotated by domain experts. However, other linguistic aspects should be considered to ensure the quality and consistency of the annotation. This paper introduces several linguistic phenomena affecting the annotation of both drug named entities and drug-drug interaction relationships that arose during the annotation process of the DDI corpus. The detailed documentation of these issues and the decisions on them will improve the quality of the annotated corpus and its usefulness for other researchers or users This work was supported by the Regional Government of Madrid under the Research Network MA2VICMR [S2009/TIC-1542] and by the Spanish Ministry of Economy under the project MULTIMEDICA [TIN2010-20644- C03-01]. Publicado

Spanish is the third-most used language on the internet, after English and Chinese, with a total of 7.7% (more than 277 million of users) and a huge internet growth of more than 1,400%. However, most work on sentiment analysis has been focused on English. This paper describes a deep learning system for Spanish sentiment analysis. To the best of our knowledge, this is the first work that explores the use of a convolutional neural network to polarity classification of Spanish tweets.

Drug-drug interaction (DDI), which is a specific type of adverse drug reaction, occurs when a drug influences the level or activity of another drug. Natural language processing techniques can provide health-care professionals with a novel way of reducing the time spent reviewing the literature for potential DDIs. The current state-of-the-art for the extraction of DDIs is based on feature-engineering algorithms (such as support vector machines), which usually require considerable time and effort. One possible alternative to these approaches includes deep learning. This technique aims to automatically learn the best feature representation from the input data for a given task. The purpose of this paper is to examine whether a convolutional neural network (CNN), which only uses word embeddings as input features, can be applied successfully to classify DDIs from biomedical texts. Proposed herein, is a CNN architecture with only one hidden layer, thus making the model more computationally efficient, and we perform detailed experiments in order to determine the best settings of the model. The goal is to determine the best parameter of this basic CNN that should be considered for future research. The experimental results show that the proposed approach is promising because it attained the second position in the 2013 rankings of the DDI extraction challenge. However, it obtained worse results than previous works using neural networks with more complex architectures. Research Program of the Ministry of Economy, Industry and Competitiveness, Government of Spain [eGovernAbility-Access project TIN2014-52665-C2-2-R].

In this paper, we describe our participation at the subtask of extraction of relationships between two identified keyphrases. This task can be very helpful in improving search engines for scientific articles. Our approach is based on the use of a convolutional neural network (CNN) trained on the training dataset. This deep learning model has already achieved successful results for the extraction relationships between named entities. Thus, our hypothesis is that this model can be also applied to extract relations between keyphrases. The official results of the task show that our architecture obtained an F1-score of 0.38% for Keyphrases Relation Classification. This performance is lower than the expected due to the generic preprocessing phase and the basic configuration of the CNN model, more complex architectures are proposed as future work to increase the classification rate.

This paper describes the system presented by the LABDA group at SemEval 2017 Task 10 ScienceIE, specifically for the subtasks of identification and classification of keyphrases from scientific articles. For the task of identification, we use the BANNER tool, a named entity recognition system, which is based on conditional random fields (CRF) and has obtained successful results in the biomedical domain. To classify keyphrases, we study the UMLS semantic network and propose a possible linking between the keyphrase types and the UMLS semantic groups. Based on this semantic linking, we create a dictionary for each keyphrase type. Then, a feature indicating if a token is found in one of these dictionaries is incorporated to feature set used by the BANNER tool. The final results on the test dataset show that our system still needs to be improved, but the conditional random fields and, consequently, the BANNER system can be used as a first approximation to identify and classify keyphrases.

Most Unified Modeling Language (UML) computer-aided software engineering tools have been insufficient in the development process because they provide little support for conceptual model testing. Model testing aims to ensure the correctness of a UML/OCL class diagram, or, in other words, that a given class diagram can perfectly meet the user's requirements. This study proposes the validation of class diagrams with black-box testing, a technique used to test software without focusing on the software's implementation or structure. An approach is proposed for the automatic transformation of the constraints of a UML/OCL class diagram into test cases. Following the creation of the test cases, they are executed with JUnit and the results produced are shown to the tester. To demonstrate the applicability of this approach, an effectiveness evaluation and an efficiency evaluation are performed here. Evaluation studies show that all faults included in a class diagram have been detected within an efficient time.

Named Entity Recognition and Classification (NERC) is an important component of applications like Opinion Tracking, Information Extraction, or Question Answering. When these applications require to work in several languages, NERC becomes a bottleneck because its development requires language-specific tools and resources like lists of names or annotated corpora. This paper presents a lightly supervised system that acquires lists of names and linguistic patterns from large raw text collections in western languages and starting with only a few seeds per class selected by a human expert. Experiments have been carried out with English and Spanish news collections and with the Spanish Wikipedia. Evaluation of NE classification on standard datasets shows that NE lists achieve high precision and reveals that contextual patterns increase recall significantly. Therefore, it would be helpful for applications where annotated NERC data are not available such as those that have to deal with several western languages or information from different domains. This researchwork has been supported by the Regional Government of Madrid under the Research Network MA2VICMR (S2009/TIC-1542), by the Spanish Ministry of Education under the project MULTIMEDICA (TIN2010-20644-C03-01) and by the Spanish Center for Industry Technological Development (CDTI, Ministry of Industry, Tourism and Trade), through the BUSCAMEDIA Project (CEN-20091026).

Ternary relationships represent the association among three entities whose constraints database designers do not always know how to manage. In other words, it is very difficult for the designer to detect, represent and add constraints in a ternary relationship according to the domain requirements. To remedy the shortcomings in capturing the semantics required for the representation of this kind of relationship, the present paper discusses a practical method to motivate the designer's use of ternary relationships in a methodological framework. The method shows how to calculate cardinality constraints in binary and ternary relationships and to preserve the associated semantics until the implementation phase of the database development method. This work forms part of the ‘Thuban: Natural Interaction Platform for Virtual Attending in Real Environments’ project (TIN2008-02711), the Spanish Ministry of Industry, Tourism and Trade through the project Semants (TSI-020100-2009-419) and also by the Spanish research projects: MA2VICMR: Improving the access, analysis and visibility of the multilingual and multimedia information in web for the Region of Madrid (S2009/TIC-1542). Publicado

A drug–drug interaction (DDI) occurs when one drug influences the level or activity of another drug. Information Extraction (IE) techniques can provide health care professionals with an interesting way to reduce time spent reviewing the literature for potential drug–drug interactions. Nevertheless, no approach has been proposed to the problem of extracting DDIs in biomedical texts. In this article, we study whether a machine learning-based method is appropriate for DDI extraction in biomedical texts and whether the results provided are superior to those obtained from our previously proposed pattern-based approach [1]. The method proposed here for DDI extraction is based on a supervised machine learning technique, more specifically, the shallow linguistic kernel proposed in Giuliano et al. (2006) [2]. Since no benchmark corpus was available to evaluate our approach to DDI extraction, we created the first such corpus, DrugDDI, annotated with 3169 DDIs. We performed several experiments varying the configuration parameters of the shallow linguistic kernel. The model that maximizes the F-measure was evaluated on the test data of the DrugDDI corpus, achieving a precision of 51.03%, a recall of 72.82% and an F-measure of 60.01%. To the best of our knowledge, this work has proposed the first full solution for the automatic extraction of DDIs from biomedical texts. Our study confirms that the shallow linguistic kernel outperforms our previous pattern-based approach. Additionally, it is our hope that the DrugDDI corpus will allow researchers to explore new solutions to the DDI extraction problem. This study was funded by the Projects MA2VICMR (S2009/TIC-1542) and MULTIMEDICA (TIN2010-20644-C03-01). Publicado

Learning content management systems (LCMSs) have become increasingly popular in the educational field over the past few years. However, problems in system design can create difficulties in the interactions between LCMSs and an important sector of the user population. The assessment and monitoring of LCMS accessibility are vital for the guarantee of universal accessibility in education. This article presents a comparative study of the accessibility of three web-based, open-source LCMSs: Moodle, ATutor, and Sakai. Results of the study indicate that barriers to accessibility are present in each of the three systems evaluated. A primary aim of the study is to help detect and correct these barriers such that the goal of universal access in educational environments may one day be achieved. © 2011 Wiley Periodicals, Inc. Comput Appl Eng Educ 22:320–328, 2014; View this article online at wileyonlinelibrary.com/journal/cae; DOI 10.1002/cae.20557

Arabic is the most widely spoken language in the Arab World. Most people of the Islamic World understand the Classic Arabic language because it is the language of the Qur'an. Despite the fact that in the last decade the number of Arabic Internet users (Middle East and North and East of Africa) has increased considerably, systems to analyze Arabic digital resources automatically are not as easily available as they are for English. Therefore, in this work, an attempt is made to build a real time Named Entity Recognition system that can be used in web applications to detect the appearance of specific named entities and events in news written in Arabic. Arabic is a highly inflectional language, thus we will try to minimize the impact of Arabic affixes on the quality of the pattern recognition model applied to identify named entities. These patterns are built up by processing and integrating different gazetteers, from DBPedia (http://dbpedia.org/About, 2009) to GATE (A general architecture for text engineering, 2009) and ANERGazet (http://users.dsic.upv.es/grupos/nle/?file=kop4.php). This work has been partially supported by the Spanish Center for Industry Technological Development (CDTI, Ministry of Industry, Tourism and Trade), through the BUSCAMEDIA Project (CEN-20091026), and also by the Spanish research projects: MA2VICMR: Improving the access, analysis and visibility of the multilingual and multimedia information in web for the Region of Madrid (S2009/TIC-1542), and MULTIMEDICA: Multilingual Information Extraction in Health domain and application to scientific and informative documents (TIN2010-20644-C03-01). The authors would like also to thank the IPSC of the European Commission’s Joint Research Centre for allowing us to include the EMM search engine in our system. Publicado

In an adaptive and intelligent educational system (AIES), the process of learning pedagogical policies according the students needs fits as a Reinforcement Learning (RL) problem. Previous works have demonstrated that a great amount of experience is needed in order for the system to learn to teach properly, so applying RL to the AIES from scratch is unfeasible. Other works have previously demonstrated in a theoretical way that seeding the AIES with an initial value function learned with simulated students reduce the experience required to learn an accurate pedagogical policy. In this paper we present empirical results demonstrating that a value function learned with simulated students can provide the AIES with a very accurate initial pedagogical policy. The evaluation is based on the interaction of more than 70 Computer Science undergraduate students, and demonstrates that an efficient and useful guide through the contents of the educational system is obtained. Publicado

16 pages, 10 figures.-- Issue title: "Best papers from the 2007 Australian Software Engineering Conference (ASWEC 2007), Melbourne, Australia, April 10-13, 2007, Australian Software Engineering Conference 2007". Transforming integrity constraints into active rules or triggers for verifying database consistency produces a serious and complex problem related to real time behaviour that must be considered for any implementation. Our main contribution to this work is to provide a complete approach for deriving the active mechanisms for Relational Databases from the specification of the integrity constraints by using OCL. This approach is designed in accordance with the MDA approach which consists of transforming the specified OCL clauses into a class diagram into SQL:2003 standard triggers, then transforming the standard triggers into target DBMS triggers. We believe that developing triggers and plugging them into a given model is insufficient because the behaviour of such triggers is invisible to the developers, and therefore not controllable. For this reason, a DBMS trigger verification model is used in our approach, in order to ensure the termination of trigger execution. Our approach is implemented as an add-in tool in Rational Rose called OCL2Trigger. This work is part of the "Software Process Management Platform: Modelling, reuse and measurement" TIN2004/07083 project. Publicado

This paper describes the participation of LABDA team in the 2016 BioASQ Task 4a on large-scale online biomedical semantic indexing. Our approach is based on the use of the open source search engine ElasticSearch. Experimental results show that our approach achieves high recall while keeping processing time low. Although more work needs to be done to improve our results, we can conclude that ElasticSearch is a competitive and scalable system for indexing biomedical literature.

Part of citizenship faces accessibility barriers when they read and understand texts containing long sentences, unusual words, complex linguistic structures, etc. Readability and understanding should be considered when texts are created. In order to make online texts more accessible, there are initiatives as Easy-to-Read (E2R) and Web Content Accessibility Guidelines (WCAG) 2.0, however they do not cover every need that arises in this regard. In addition to accessibility guidelines and E2R rules, technology supporting the authorship of texts is required because the transformation of a text to an easier text to read and understandable is impossible. As a solution to the need to have (semi)automatic support to comply with accessibility and E2R guidelines, the application of Natural Language Processing (NLP) resources and methods is proposed. In this paper, E2R guidelines are introduced and a subset of WCAG guidelines regarding readability and understanding has been obtained. In addition, a review on NLP technology concerning accessibility is given. To illustrate this proposal for improving accessibility using PLN approaches, a use case for simplifying drug package leaflets in Spanish is introduced.

The early detection of drug drug interactions (DDIs) is limited by the diffuse spread of DDI information in heterogeneous sources. Computational methods promise to play a key role in the identification and explanation of DDIs on a large scale. However, such methods rely on the availability of computable representations describing the relevant domain knowledge. Current modeling efforts have focused on partial and shallow representations of the DDI domain, failing to adequately support computational inference and discovery applications. In this paper, we describe a comprehensive ontology for DDI knowledge (DINTO), which is the first formal representation of different types of DDIs and their mechanisms and its application in the prediction of DDIs. This project has been developed using currently available semantic web technologies, standards, and tools, and we have demonstrated that the combination of drug-related facts in DINTO and Semantic Web Rule Language (SWRL) rules can be used to infer DDIs and their different mechanisms on a large scale.< This work was supported by the European Commission Seventh Framework Programme under the TrendMiner project [FP7-ICT287863].

28 pages, 13 figures.-- Special Issue: "Symbiotic Performance between Humans and Intelligent Systems". This paper presents an interaction model pursuing flexible and coherent human–computer interaction. Starting from a cognitive architecture for Natural Interaction, an agent-based design is presented, focusing particularly on the role of the interaction agent. Regarding the intentional processing within this agent, the Threads Model is proposed. Finally, its implementation is described and evaluated to find out the integrity of the intentional approach. We thank the ISYS research group at the Technical University of Madrid and the LABDA research group at the Carlos III University of Madrid for their support during the design and development of the interaction agent, and everyone involved in the ADVICE (IST 1999-11305), VIP-ADVISOR (IST 2001-32440), and recent IntegraTV4All (FIT-350301-2004-2) projects for their hard work.

This paper presents our work in the ADVICE (IST-1999-11305, Information Societies Technology Programme, V Framework Programme of the European Commission) project, an on-going European Commission research project of the IST Programme. The overall objective of this project is to design and implement an advice-giving system for E-commerce, supporting a move from the current catalogue-based customer services to a customer adapted intelligent assistance, emulating in some way the performance of a human seller. With this aim, the main elements of the ADVICE approach include an agent-based architecture with an Interface Agent to manage the multimedia presentation, the Interaction Agent to support an advanced user–system interaction and the Intelligent Agent incorporating a knowledge-based model of the e-business for the pilot domain, that supports the reasoning for advise-giving according with the user needs and the dialogue evolution.

To apply Extended Entity Relationship Model (EER) is a good method for representing requirements on information systems, because of its high level of abstraction. Although it is very close to the user, it is not so trivial when some constructs, such as higher order relationships, are used. This paper describes the characterisation and several important results of an experiment performed at our university in order to show some of the difficu lties found when novice students and practitioners use ternary relationships. Some special topics in identifying ternary relationships such as the importance of the domain of text and the intersection data are also investigated. In order to guide and help users in the design task, these results are introduced in PANDORA Case Tool, a research project which tries to serve as a methodological assistance tool.

Despite the myriad media content, there are users who cannot access it. Accessibility standards indicate that both the media content as the user agent that provides access to the media content must be accessible to everyone, with or without a disability. To address this issue, this work presents a Model-Based tool, which gives support to the development process of an accessible media player according to accessibility standards. Requirements and adaptation rules regarding the accessibility of a media player and video content are included in the tool design following a methodological approach. This authoring tool is a graphical editor which is aimed at being used by any professional without having a high level of expertise in accessibility.

This paper describes a machine learningbased approach that uses word embedding features to recognize drug names from biomedical texts. As a starting point, we developed a baseline system based on Conditional Random Field (CRF) trained with standard features used in current Named Entity Recognition (NER) systems. Then, the system was extended to incorporate new features, such as word vectors and word clusters generated by the Word2Vec tool and a lexicon feature from the DINTO ontology. We trained the Word2vec tool over two different corpus: Wikipedia and MedLine. Our main goal is to study the effectiveness of using word embeddings as features to improve performance on our baseline system, as well as to analyze whether the DINTO ontology could be a valuable complementary data source integrated in a machine learning NER system. To evaluate our approach and compare it with previous work, we conducted a series of experiments on the dataset of SemEval-2013 Task 9.1 Drug Name Recognition.

The automatic extraction of chemical information from text requires the recognition of chemical entity mentions as one of its key steps. When developing supervised named entity recognition (NER) systems, the availability of a large, manually annotated text corpus is desirable. Furthermore, large corpora permit the robust evaluation and comparison of different approaches that detect chemicals in documents. We present the CHEMDNER corpus, a collection of 10,000 PubMed abstracts that contain a total of 84,355 chemical entity mentions labeled manually by expert chemistry literature curators, following annotation guidelines specifically defined for this task. The abstracts of the CHEMDNER corpus were selected to be representative for all major chemical disciplines. Each of the chemical entity mentions was manually labeled according to its structure-associated chemical entity mention (SACEM) class: abbreviation, family, formula, identifier, multiple, systematic and trivial. The difficulty and consistency of tagging chemicals in text was measured using an agreement study between annotators, obtaining a percentage agreement of 91. For a subset of the CHEMDNER corpus (the test set of 3,000 abstracts) we provide not only the Gold Standard manual annotations, but also mentions automatically detected by the 26 teams that participated in the BioCreative IV CHEMDNER chemical mention recognition task. In addition, we release the CHEMDNER silver standard corpus of automatically extracted mentions from 17,000 randomly selected PubMed abstracts. A version of the CHEMDNER corpus in the BioC format has been generated as well. We propose a standard for required minimum information about entity annotations for the construction of domain specific corpora on chemical and drug entities. The CHEMDNER corpus and annotation guidelines are available at: http://www.biocreative.org/resources/biocreative-iv/chemdner-corpus/ This work is supported by the Innovative Medicines Initiative Joint Undertaking (IMI-eTOX) and the MICROME grant 222886-2.

The concept of multiplicity in UML derives from that of cardinality in entity-relationship modeling techniques. The UML documentation defines this concept but at the same time acknowledges some lack of obviousness in the specification of multiplicities for n-ary associations. This paper shows an ambiguity in the definition given by UML documentation and proposes a clarification to this definition, as well as the use of outer and inner multiplicities as a simple extension to the current notation to represent other multiplicity constraints, such as participation constraints, that are equally valuable in understanding n-ary associations.

Background: Biomedical semantic indexing is a very useful support tool for human curators in their efforts for indexing and cataloging the biomedical literature. Objective: The aim of this study was to describe a system to automatically assign Medical Subject Headings (MeSH) to biomedical articles from MEDLINE. Methods: Our approach relies on the assumption that similar documents should be classified by similar MeSH terms. Although previous work has already exploited the document similarity by using a k-nearest neighbors algorithm, we represent documents as document vectors by search engine indexing and then compute the similarity between documents using cosine similarity. Once the most similar documents for a given input document are retrieved, we rank their MeSH terms to choose the most suitable set for the input document. To do this, we define a scoring function that takes into account the frequency of the term into the set of retrieved documents and the similarity between the input document and each retrieved document. In addition, we implement guidelines proposed by human curators to annotate MEDLINE articles; in particular, the heuristic that says if 3 MeSH terms are proposed to classify an article and they share the same ancestor, they should be replaced by this ancestor. The representation of the MeSH thesaurus as a graph database allows us to employ graph search algorithms to quickly and easily capture hierarchical relationships such as the lowest common ancestor between terms. Results: Our experiments show promising results with an F1 of 69% on the test dataset. Conclusions: To the best of our knowledge, this is the first work that combines search and graph database technologies for the task of biomedical semantic indexing. Due to its horizontal scalability, ElasticSearch becomes a real solution to index large collections of documents (such as the bibliographic database MEDLINE). Moreover, the use of graph search algorithms for accessing MeSH information could provide a support tool for cataloging MEDLINE abstracts in real time.

The constant increase of multimedia content like video on the Web does not entail an equitable increase of the accessibility of this content. This fact provokes that elderly people or people with some kind of disability are excluded of the access to this Web content. This exclusion is due to the fact that the Web user agent that provides the content does not offer alternative content that allows users with disabilities to access it. In the case of a media player, this user agent has to provide control for access this alternative content such as caption or audio description. So, the access to alternative content should be assured to the user with disabilities in accordance with their access needs. Because of these facts, the motivation of this research work arises. The contribution of this research work aims to provide a model-based design solution which includes adaptation rules for accessible media player following accessibility standards.

Studies that have been done for more than a decade indicate a persistent shortage of women's participation in STEM studies (Science, Technology, Engineering and Mathematics). Women do not usually choose STEM studies, and their participation in the ICT business sector is still lower. The Polytechnic School (EPS) of the Carlos III University of Madrid (UC3M) has carried out a survey study with the aim of collecting data to understand the perception of students/teachers, and analyze the factors of the gender gap. The conclusions of the study are introduced.

It is a legal requirement in many countries to ensure accessibility for Web applications. Although there are numerous regulations and standards regarding accessibility in the development of user interfaces, accessibility is nevertheless quite difficult to achieve, let alone to maintain at a high level of quality throughout the development process. This difficulty is due to diverse factors including, but not limited to, the lack of proper development methods, authoring tools and accessibility training for user interface development professionals. In an attempt to offer a solution to these difficulties, this paper proposes a methodological approach for the integration of accessibility requirements into a user interface development method based on User Interface Description Language (UIDL) or, more specifically, on the USer Interface eXtensible Markup Language (UsiXML) framework. The proposed strategy involves the integration of accessibility requirements into design primitives of the user interface development method. This paper focuses on accessibility requirements related to navigation in the Task Model, Abstract User Interface Model and Transformation Model. The application of the approach shown for the SPA4USXML tool also includes a proof of concept and validation of the proposal. The study shows that accessibility requirements included at the design stage in the modelling of user interfaces can be systematized through mechanisms such as new transformation rules and the use of support tools such as SPA4USXML. As a result, the quality of accessibility can be guaranteed by taking advantage of developments in technologies based on User Interface Markup Languages such as UsiXML. This research work has been partly funded by the Department of Education, Universities and Research of the Basque Government and by the Research Network MAVIR (S2009/TIC-1542 (www.mavir.net), and MULTIMEDICA Project (TIN2010- 20644-C03-01). In addition, RM enjoys a PhD scholarship from the Research Staff Training Programme of the Department of Education, Universities and Research of the Basque Government. We thank the INREDIS project, which has been the foundation for this work. We also want to thank the members of the ITEA Project for providing us with the documents of the UsiXML specification Publicado