Search

Nowadays, most health professionals use electronic health records to keep track of patients. To properly use and share these data, the community has relied on medical classification standards to represent patient information. However, the coding process is tedious and time-consuming, often limiting its application. This paper proposes a novel feature representation method that considers the distinction between diagnoses and procedure codes, and applies this to the task of medical procedure code prediction. Diagnosis codes are combined with text annotations, and the result is then used as input to a downstream procedure code prediction task. Various diagnosis code representations are considered by exploiting a code hierarchy. Furthermore, different text representation strategies are also used, including embeddings from language models. Finally, the method was evaluated using the MIMIC-III database. Our experiments showed improved performance in procedure code prediction when exploiting the diagnosis codes, outperforming state-of-the-art models.

Green areas play an important role in people’s well-being in urban areas. However, traditional survey methods hinder understanding their actual impact. Fortunately, social networking analysis provides valuable information that city planners can use to transform cities and improve city life. This research studies geolocated tweets published in parks, both urban and natural, in Madrid, for their subsequent analysis and classification with machine-learning techniques, and determines the emotional impact of green areas on citizens. The main conclusions of this study are that people express more positive sentiments and emotions (i.e., joy and trust) in urban parks in Madrid compared to the sentiments expressed in other areas of the city and a national park in the Madrid region. This positive sentiment is higher in the city’s southern districts and the historical parks. People also tweeted photos more frequently in parks and differences in the topics expressed in the tweets. This analysis can provide additional information to policymakers in urban planning.

Depression is a common and debilitating mental illness affecting millions of individuals, diminishing their quality of life and overall well-being. The increasing prevalence of mental health disorders has underscored the need for innovative approaches to detect and address depression. In this context, text analysis has emerged as a promising avenue. Novel solutions for text-based depression detection commonly rely on deep neural networks or transformer-based models. Although these approaches have yielded impressive results, they often come with inherent limitations, such as substantial computational requirements or a lack of interpretability. This work aims to bridge the gap between substantial performance and practicality in the detection of depression signs within digital content. To this end, we introduce a comprehensive feature framework that integrates linguistic signals, emotional expressions, and cognitive patterns. The combination of this framework with distributional representations contributes to fostering the understanding of language patterns indicative of depression and provides a deeper grasp of contextual nuances. We exploit this combination using traditional machine learning methods in an effort to yield substantial performance without compromising interpretability and computational efficiency. The performance and generalizability of our approach have been assessed through experimentation using multiple publicly available English datasets. The results demonstrate that our method yields throughput on par with more complex and resource-intensive solutions, achieving F1-scores above 70%. This accomplishment is notable, as the proposed method simultaneously preserves the virtues of simplicity, interpretability, and reduced computational overhead. In summary, the findings of this research contribute to the field by offering an accessible and scalable solution for the detection of depression in real-world scenarios.

Whether consciously or inadvertently, our messages can include toxic language which contributes to the polarization of social networks. Intelligent techniques can help us detect these expressions and even change them into kinder expressions by applying style transfer techniques. This work aims to advance detoxification style transfer techniques using deep learning and semantic similarity technologies. The article explores the advantages of a toxicity-deletion method that uses linguistic resources in a detoxification system. For this purpose, we propose a method that removes toxic words from the source sentence using a similarity function with a toxic vocabulary. We present two models that leverage it, namely, LexiconGST and MultiLexiconGST, which are based on the Delete –Retrieve–Generate framework. Experimental results show that our models perform well in the detoxification task compared to other state-of-the-art methods. Finally, this research confirms that linguistic resources can guide deep learning techniques and improve their performance.

Interest in social media has only increased with time [...]

The Internet has become an important tool for modern terrorist groups as a means of spreading their propaganda messages and recruitment purposes. Previous studies have shown that the analysis of social signs can help in the analysis, detection, and prediction of radical users. In this work, we focus on the analysis of affect signs in social media and social networks, which has not been yet previously addressed. The article contributions are: (i) a novel dataset to be used in radicalization detection works, (ii) a method for utilizing an emotion lexicon for radicalization detection, and (iii) an application to the radical detection domain of an embedding-based semantic similarity model. Results show that emotion can be a reliable indicator of radicalization, as well as that the proposed feature extraction methods can yield high-performance scores.

The application of natural language to improve students' interaction with information systems is demonstrated to be beneficial. In particular, advances in cognitive computing enable a new way of interaction that accelerates insight from existing information sources, thereby contributing to the process of learning. This work aims at researching the application of cognitive computing in blended learning environments. We propose a modular cognitive agent architecture for pedagogical question answering, featuring social dialogue (small talk), improved for a specific knowledge domain. This system has been implemented as a personal agent to assist students in learning Data Science and Machine Learning techniques. Its implementation includes the training of machine learning models and natural language understanding algorithms in a human-like interface. The effectiveness of the system has been validated through an experiment.

GSITK is a framework to perform a wide variety of sentiment analysis tasks, including dataset acquisition, text preprocessing, model design, and performance evaluation. The framework is oriented to both researchers and practitioners, easing the replication of previous sentiment models, as well as offering implementations of common tasks. This is achieved by building several abstractions on top of popular libraries such as scikit-learn and NLTK. In this way, GSITK allows users to implement complex sentiment pipelines using comprehensible Python code. The framework is Open Source and has been used successfully in several research projects and competitions.

Understanding radicalization pathways, drivers, and factors is essential for the effective design of prevention and counter-radicalization programs. Traditionally, the primary methods used by social scientists to detect these drivers and factors include literature reviews, qualitative interviews, focus groups, and quantitative methods based on surveys. This article proposes to complement social science approaches with computational methods to detect these factors automatically by analyzing the language signals expressed in social networks. To this end, the article categorizes radicalization drivers and factors following the micro, meso, and macro levels used in the social sciences. It identifies the corresponding language signals and available language resources. Then, a computational system is developed to monitor these language signals. In addition, this article proposes semantic technologies since they offer unique exploration, query, and discovery capabilities. The system was evaluated based on a set of competency questions that show the benefits of this approach.

The recent increase in user interaction with social media has completely changed the way customers communicate their opinions, questions, and concerns to brands. For this reason, many companies have established on the top of their agendas the necessity of analyzing the high amounts of user-generated content data in social networks. These analyses are helping brands to understand their customers’ experiences as well as for maintaining a competitive advantage in the sector. Due to this fact, this study aims to analyze and characterize the public opinions from the messages posted by Twitter users while addressing customer services. For this purpose, this study carried out a content analysis of a customer service platform. We extracted the general users’ viewpoints and sentiments of each of the discussed topics by using a wide range of techniques, such as topic modeling, document clustering, and opinion mining algorithms. For training these systems and drawing conclusions, a dataset containing tweets from the English-speaking customers addressing the @Uber_Support platform during the year 2020 has been used.

Nowadays, we are witnessing a shift in the way emergencies are being managed. On the one hand, the availability of big data and the evolution of geographical information systems make it possible to manage and process large quantities of information that can hugely improve the decision-making process. On the other hand, digital humanitarianism has shown to be very beneficial for providing support during emergencies. Despite this, the full potential of combining automatic big data processing and digital humanitarianism approaches has not been fully realized, though there is an initial body of research. This paper aims to provide a reference architecture for emergency management that instantiates the NIST Big Data Reference Architecture to provide a common language and enable the comparison of solutions for solving similar problems.

Museums play a crucial role in preserving cultural heritage. However, the forms in which they display cultural heritage might not be the most effective at piquing visitors’ interest. Therefore, museums tend to integrate different technologies that aim to create engaging and memorable experiences. In this context, the emerging Internet of Things (IoT) technology results particularly promising due to the possibility of implementing smart objects in museums, granting exhibits advanced interaction capabilities. Gamification techniques are also a powerful technique to draw visitors’ attention. These often rely on interactive question-based games. A drawback of such games is that questions must be periodically regenerated, and this is a time-consuming task. To confront these challenges, this paper proposes a low-maintenance gamified smart object platform that automates the creation of questions by exploiting semantic web technologies. The platform has been implemented in a real-life scenario. The results obtained encourage the use of the platform in the museum considered. Therefore, it appears to be a promising work that could be extrapolated and adapted to other kinds of museums or cultural heritage institutions.

In recent years, the sharing economy has become popular, with outstanding examples such as Airbnb, Uber, or BlaBlaCar, to name a few. In the sharing economy, users provide goods and services in a peer-to-peer scheme and expose themselves to material and personal risks. Thus, an essential component of its success is its capability to build trust among strangers. This goal is achieved usually by creating reputation systems where users rate each other after each transaction. Nevertheless, these systems present challenges such as the lack of information about new users or the reliability of peer ratings. However, users leave their digital footprints on many social networks. These social footprints are used for inferring personal information (e.g., personality and consumer habits) and social behaviors (e.g., flu propagation). This article proposes to advance the state of the art on reputation systems by researching how digital footprints coming from social networks can be used to predict future behaviors on sharing economy platforms. In particular, we have focused on predicting the reputation of users in the second-hand market Wallapop based solely on their users’ Twitter profiles. The main contributions of this research are twofold: (a) a reputation prediction model based on social data; and (b) an anonymized dataset of paired users in the sharing economy site Wallapop and Twitter, which has been collected using the user self-mentioning strategy.

Recent works have shown that sentiment analysis on social media can be improved by fusing text with social context information. Social context is information such as relationships between users and interactions of users with content. Although existing works have already exploited the networked structure of social context by using graphical models or techniques such as label propagation, more advanced techniques from social network analysis remain unexplored. Our hypothesis is that these techniques can help reveal underlying features that could help with the analysis. In this work, we present a sentiment classification model (CRANK) that leverages community partitions to improve both user and content classification. We evaluated this model on existing datasets and compared it to other approaches.

Sentiment analysis has become a key technology to gain insight from social networks. The field has reached a level of maturity that paves the way for its exploitation in many different fields such as marketing, health, banking or politics. The latest technological advancements, such as deep learning techniques, have solved some of the traditional challenges in the area caused by the scarcity of lexical resources. In this Special Issue, different approaches that advance this discipline are presented. The contributed articles belong to two broad groups: technological contributions and applications.

In the framework of the Spectral Line Shapes in Plasmas Code Comparison Workshop (SLSP), large discrepancies appeared between the different approaches to account for ion motion effects in spectral line shape calculations. For a better understanding of these effects, in the second edition of the SLSP in August, 2013, two cases were dedicated to the study of the ionic field directionality on line shapes. In this paper, the effects of the direction and magnitude fluctuations are separately analyzed. The effects of two variants of electric field models, (i) a pure rotating field with constant magnitude and (ii) a time-dependent magnitude field in a given direction, together with the effects of the time-dependent ionic field on shapes of the He II Lyman-α and -β lines for different densities and temperatures, are discussed.

Traditionally, fault diagnosis in telecommunication network management is carried out by humans who use software support systems. The phenomenal growth in telecommunication networks has nonetheless triggered the interest in more autonomous approaches, capable of coping with emergent challenges such as the need to diagnose faults’ root causes under uncertainty in geographically-distributed environments, with restrictions on data privacy. In this paper, we present a framework for distributed fault diagnosis under uncertainty based on an argumentative framework for multi-agent systems. In our approach, agents collaborate to reach conclusions by arguing in unpredictable scenarios. The observations collected from the network are used to infer possible fault root causes using Bayesian networks as causal models for the diagnosis process. Hypotheses about those fault root causes are discussed by agents in an argumentative dialogue to achieve a reliable conclusion. During that dialogue, agents handle the uncertainty of the diagnosis process, taking care of keeping data privacy among them. The proposed approach is compared against existing alternatives using benchmark multi-domain datasets. Moreover, we include data collected from a previous fault diagnosis system running in a telecommunication network for one and a half years. Results show that the proposed approach is suitable for the motivational scenario.

Several GWAS reported Myocyte Enhancer Factor 2 C (MEF2C) gene associations with white matter microstructure and psychiatric disorders, and MEF2C involvement in pathways related to neuronal development suggests a common biological factor underlying these phenotypes. We aim to refine the MEF2C effects in the brain relying on an integrated analysis of white matter and psychiatric phenotypes in an extensively characterized sample. This study included 870 Brazilian adults (47% from an attention-deficit/hyperactivity disorder outpatient clinic) assessed through standardized psychiatric interviews, 139 of which underwent a magnetic resonance imaging scan. We evaluated variants in the MEF2C region using two approaches: 1) a gene-wide analysis, which uses the sum of polymorphism effects, and 2) SNP analyses, restricted to the independent variants within the gene. The outcomes included psychiatric phenotypes and fractional anisotropy for brain images. Results: The gene-wide analyses pointed to a nominal association between MEF2C and the Temporal Portion of the Superior Longitudinal Fasciculus (SLFTEMP). The SNP analysis identified four independent variants significantly associated with SLFTEMP and one (rs4218438) with Substance Use Disorder. Our findings showing specific associations of MEF2C variants with temporal-frontal circuitry components may help to elucidate how the MEF2C gene underlies a broad range of psychiatric phenotypes since these regions are relevant to executive and cognitive functions.

Este artículo presenta la plataforma TutorGSI que facilita el uso de agentes conversacionales en entornos de eLearning como Moodle. La plataforma ha sido desarrollada dentro del proyecto eduWAI (eduWAI, 2010), y simplifica el proceso de resolución de dudas académicas a través de un agente con el que se interactúa en lenguaje natural. TutorGSI posibilita que los profesores puedan modificar las conversaciones existentes o crear nuevos tipos en función de las consultas que vayan planteando los alumnos.

n/a

The evolution of the Internet of Things leads to new opportunities for the contemporary notion of smart offices, where employees can benefit from automation to maximize their productivity and performance. However, although extensive research has been dedicated to analyze the impact of workers’ emotions on their job performance, there is still a lack of pervasive environments that take into account emotional behaviour. In addition, integrating new components in smart environments is not straightforward. To face these challenges, this article proposes an architecture for emotion aware automation platforms based on semantic event-driven rules to automate the adaptation of the workplace to the employee’s needs. The main contributions of this paper are: (i) the design of an emotion aware automation platform architecture for smart offices; (ii) the semantic modelling of the system; and (iii) the implementation and evaluation of the proposed architecture in a real scenario.

We report experimental determination of plasma-induced Stark widths and shifts of the He II P α line and a comparison of the results with calculations performed by several computational approaches. The measurements were carried out in a small compressing plasma channel device, reaching electron densities in excess of 10 18 cm − 3 and temperatures of a few eV. The experimental data are in a good agreement with some previously published studies. However, the measured relation between the Stark shift and width could not be reproduced by either of the codes, and this disagreement is not yet resolved. This suggests the existence of an additional effect that is not accounted for in the present models and leads to a larger than expected Stark shift of the He II P α line.

Indoor evacuation systems are needed for rescue and safety management. One of the challenges is to provide users with personalized evacuation routes in real time. To this end, this project aims at exploring the possibilities of Google Glass technology for participatory multiagent indoor evacuation simulations. Participatory multiagent simulation combines scenario-guided agents and humans equipped with Google Glass that coexist in a shared virtual space and jointly perform simulations. The paper proposes an architecture for participatory multiagent simulation in order to combine devices (Google Glass and/or smartphones) with an agent-based social simulator and indoor tracking services.

Background The Russia-Ukraine war brought immediate and delayed socio-economic consequences. In general, the repercussions caused fear all over the world. This study aims to validate an instrument for measuring fear perception caused by the consequences of a large-scale war in Latin American citizens. Methodology An instrumental study in which 1705 residents of Bolivia, Colombia, Ecuador, Panama, Paraguay, Peru and other countries were surveyed through a virtual format. A literature search, expert judgment, preliminary (then exploratory and confirmatory) analysis, as well as reliability assessment were carried out. Results The skewness and kurtosis values of the 13 questions did not exceed the range ± 1.5 and showed significant correlations (>0.30). The Kaiser-Meyer-Olkin index (0.962) and Bartlett's test (19558.5; df=78; p=0.001) had good indicators. The parallel analysis suggested a single factor, which explained 75.59% of the total variance. The confirmatory factor analysis generated an instrument with six items (χ 2=47.33, df=9, p=0.001; RMR=0.010; GFI=0.990; CFI=1.00; TLI=0.990; and RMSEA=0.050), with an overall Cronbach's Alpha=0.949 (95% CI=0.94–0.95). Conclusion A six-item instrument that measures the perception of fear caused by the consequences of a large-scale war was validated in half a dozen Latin American countries. This short and valid instrument can be administered to a broad population in Latin America.