publications | Alejandro Benito-Santos

2026

Chasing Meaning and/or Insight? A Survey on Evaluation Practices at the Intersection of Visualization and the Humanities

Alejandro Benito-Santos, Florian Windhager, Aida Horaniet Ibañez, and 3 more authors

In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems, Barcelona, Spain, Apr 2026

Best Paper Award Abs DOI arXiv PDF

Best Paper Award at CHI 2026.

We surveyed 171 visualization design studies in the humanities to examine how their evaluation practices address both analytical utility and interpretive meaning. The study identifies recurring methodological weaknesses and recommends combining diverse forms of evidence to evaluate how visualizations support humanities research.
The Ruler and the Judge: Benchmark-Conditional Evaluation of LLM-as-a-Judge

Adrián Ghajari, Alejandro Benito-Santos, and Víctor Fresno

In 40th Conference on Neural Information Processing Systems (NeurIPS 2026), Evaluations and Datasets Track, 2026

Accepted, to appear

PDF

2025

On the Correspondence between the Squared Norm and Information Content in Text Embeddings

Enrique Amigo, Adrian Ghajari, Alejandro Benito-Santos, and 1 more author

In Findings of the Association for Computational Linguistics: EMNLP 2025, Nov 2025

Abs DOI PDF

Previous work has reported both empirical and theoretical evidence, for specific training models, of the correspondence between the squared norm of an embedding and the information content of the text it represents.In this paper, we investigate the relationship at the theoretical and empirical levels, focusing on the mechanisms and composition functions used to combine token embeddings. i) We formally derive two sufficient theoretical conditions for this correspondence to hold in embedding models. ii) We empirically examine the correspondence and the validity of these conditions at the word level for both static and contextual embeddings and different subword token composition mechanisms.iii) Building on Shannon’s Constant Entropy Rate (CER) principle, we explore whether embedding mechanisms exhibit a linearly monotonic increase in information content as text length increases.Our formal analysis and experiments reveal that:i) At the word embedding level, models satisfy the sufficient conditions and show a strong correspondence when certain subword composition functions are applied.ii) Only scaled embedding averages proposed in this paper and certain information-theoretic composition functions preserve the correspondence. Some non-compositional representations—such as the CLS token in BERT or the EOS token in LLaMA—tend to converge toward a fixed point. The CLS token in ModernBERT, however, exhibits behavior that aligns more closely with the CER hypothesis.
Beyond Averages: Learning with Annotator Disagreement in STS

Alejandro Benito-Santos and Adrian Ghajari

In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, Nov 2025

Abs PDF

This work investigates capturing and modeling disagreement in Semantic Textual Similarity (STS), where sentence pairs are assigned ordinal similarity labels (0–5). Conventional STS systems average multiple annotator scores and focus on a single numeric estimate, overlooking label dispersion. By leveraging the disaggregated SemEval-2015 dataset (Soft-STS-15), this paper proposes and compares two disagreement-aware strategies that treat STS as an ordinal distribution prediction problem: a lightweight truncated Gaussian head for standard regression models, and a cross-encoder trained with a distance-aware objective, refined with temperature scaling. Results show improved performance in distance-based metrics, with the calibrated soft-label model proving best overall and notably more accurate on the most ambiguous pairs. This demonstrates that modeling disagreement benefits both calibration and ranking accuracy, highlighting the value of retaining and modeling full annotation distributions rather than collapsing them to a single mean label.
Robust Estimation of Population-Level Effects in Repeated-Measures NLP Experimental Designs

Alejandro Benito-Santos, Adrian Ghajari, and Víctor Fresno

In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2025

Abs PDF

NLP research frequently grapples with multiple sources of variability—spanning runs, datasets, annotators, and more—yet conventional analysis methods often neglect these hierarchical structures, threatening the reproducibility of findings. To address this gap, we contribute a case study illustrating how linear mixed-effects models (LMMs) can rigorously capture systematic language-dependent differences (i.e., population-level effects) in a population of monolingual and multilingual language models. In the context of a bilingual hate speech detection task, we demonstrate that LMMs can uncover significant population-level effects—even under low-resource (small-N) experimental designs—while mitigating confounds and random noise. By setting out a transparent blueprint for repeated-measures experimentation, we encourage the NLP community to embrace variability as a feature, rather than a nuisance, in order to advance more robust, reproducible, and ultimately trustworthy results.
BOE-XSUM: Extreme Summarization in Clear Language of Spanish Legal Decrees and Notifications

Andrés Fernández García, Javier Rosa, Julio Gonzalo, and 9 more authors

Procesamiento del Lenguaje Natural, Sep 2025

Abs PDF

The ability to summarize long documents succinctly is increasingly important in daily life due to information overload, yet there is a notable lack of such summaries for Spanish documents in general, and in the legal domain in particular. In this work, we present BOE-XSUM, a curated dataset comprising 3,648 concise, plain-language summaries of documents sourced from Spain’s "Boletín Oficial del Estado" (BOE), the State Official Gazette. Each entry in the dataset includes a short summary, the original text, and its document type label. We evaluate the performance of medium-sized large language models (LLMs) fine-tuned on BOEXSUM, comparing them to general-purpose generative models in a zero-shot setting. Results show that fine-tuned models significantly outperform their non-specialized counterparts. Notably, the best-performing model—BERTIN GPT-J 6B (32-bit precision)—achieves a 24% performance gain over the top zero-shot model, DeepSeek-R1 (accuracies of 41.6% vs. 33.5%).
Test-Driving Information Theory-Based Compositional Distributional Semantics: A Case Study on Spanish Song Lyrics

Adrián Ghajari, Alejandro Benito-Santos, Salvador Ros, and 2 more authors

Knowledge-Based Systems, Jun 2025

Abs DOI PDF

Song lyrics pose unique challenges for semantic similarity assessment due to their metaphorical language, structural patterns, and cultural nuances - characteristics that often challenge standard natural language processing (NLP) approaches. These challenges stem from a tension between compositional and distributional semantics: while lyrics follow compositional structures, their meaning depends heavily on context and interpretation. The Information Theory-based Compositional Distributional Semantics framework offers a principled approach by integrating information theory with compositional rules and distributional representations. We evaluate eight embedding models on Spanish song lyrics, including multilingual, monolingual contextual, and static embeddings. Results show that multilingual models consistently outperform monolingual alternatives, with the domain-adapted ALBERTI achieving the highest F1 macro scores (78.92 \textpm 10.86). Our analysis reveals that monolingual models generate highly anisotropic embedding spaces, significantly impacting performance with traditional metrics. The Information Contrast Model metric proves particularly effective, providing improvements up to 18.04 percentage points over cosine similarity. Additionally, composition functions maintaining longer accumulated vector norms consistently outperform standard averaging approaches. Our findings have important implications for NLP applications and challenge standard practices in similarity calculation, showing that effectiveness varies with both task nature and model characteristics.
Bilingual Evaluation of Language Models on General Knowledge in University Entrance Exams with Minimal Contamination

Eva Sánchez Salido, Roser Morante, Julio Gonzalo, and 8 more authors

In Proceedings of the 31st International Conference on Computational Linguistics, Jan 2025

Abs PDF

In this article we present UNED-ACCESS 2024, a bilingual dataset that consists of 1003 multiple-choice questions of university entrance level exams in Spanish and English. Questions are originally formulated in Spanish and manually translated into English, and have not ever been publicly released, ensuring minimal contamination when evaluating Large Language Models with this dataset. A selection of current open-source and proprietary models are evaluated in a uniform zero-shot experimental setting both on the UNED-ACCESS 2024 dataset and on an equivalent subset of MMLU questions. Results show that (i) Smaller models not only perform worse than the largest models, but also degrade faster in Spanish than in English. The performance gap between both languages is negligible for the best models, but grows up to 37% for smaller models; (ii) Model ranking on UNED-ACCESS 2024 is almost identical (0.98 Pearson correlation) to the one obtained with MMLU (a similar, but publicly available benchmark), suggesting that contamination affects similarly to all models, and (iii) As in publicly available datasets, reasoning questions in UNED-ACCESS are more challenging for models of all sizes.

2023

LyricSIM: Un nuevo dataset y benchmark para la detección de similitud en letras de canciones en español

Alejandro Benito Santos, Adrián Ghajari Espinosa, Pedro Hernández, and 3 more authors

Procesamiento del lenguaje natural, 2023

Abs PDF

En este trabajo presentamos un nuevo conjunto de datos y benchmark orientados a la tarea de similitud semántica en letras de canciones. Nuestro conjunto de datos, originalmente formado por 2775 pares de canciones en Español, fue anotado en un experimento de anotación colectivo por 63 anotadores nativos. Después de recoger y refinar los datos para asegurar un alto grado de consenso e integridad en los datos, obtuvimos 676 pares anotados de alta calidad que fueron empleados para evaluar el rendimiento de diferentes modelos del lenguaje monolingües y multilingües pertenecientes al estado del arte. En consecuencia, obtuvimos unos resultados base que esperamos sean de utilidad a la comunidad en todas aquellas aplicaciones académicas e industriales futuras que se realicen en este contexto. In this paper, we present a new dataset and benchmark tailored to the task of semantic similarity in song lyrics. Our dataset, originally consisting of 2775 pairs of Spanish songs, was annotated in a collective annotation experiment by 63 native annotators. After collecting and refining the data to ensure a high degree of consensus and data integrity, we obtained 676 high-quality annotated pairs that were used to evaluate the performance of various state-of-the-art monolingual and multilingual language models. Consequently, we established baseline results that we hope will be useful to the community in all future academic and industrial applications conducted in this context.
Characterizing the Visualization Design Space of Distant and Close Reading of Poetic Rhythm

Alejandro Benito-Santos, Salvador Muñoz, Roberto Therón Sánchez, and 1 more author

Frontiers in Big Data, 2023

Abs PDF

Metrical and rhythmical poetry analysis is founded on the systematic statistical analysis and comparison of sonic devices (e.g., rhythmic patterns) that emerge from a combination of pre-established aesthetic and structural rules and the poet’s abilities and creative genius to convey a given message adhering to the said constraints. These rhythmical patterns, which have been traditionally obtained by means of a careful close reading of the poems, in a process known as “scansion,” can now be obtained and made visible by automatic means. However, the visualization literature is still scarce on approaches that allow an insightful close and distant reading of the rhythmical patterns in a poetry corpus. In this work, we report our initial efforts in characterizing of the visualization design space of distant and close reading of poetic rhythm. By employing a digital version of a corpus of 11,268 verses originally written by the Spanish poet and playwright Federico García-Lorca (1898–1936), we could craft several prototypical visualizations representative of the inherent complexity of the problem which we expect to employ in future user studies and that we share here with the rest of the community to foster further discussion around this interesting topic.
Making the Whole Greater than the Sum of Its Parts: Taxonomy Development as a Site of Negotiation and Compromise in an Interdisciplinary Software Development Project

Jennifer C. Edmond, Alejandro Benito Santos, Michelle Doran, and 5 more authors

Digital Humanities Quarterly, Jul 2023

PDF

2021

Playing Design: A Case Study on Applying Gamification to Construct a Serious Game with Youngsters at Social Risk

Alejandro Benito-Santos, Amelie Dorn, Antonio G. Losada Gómez, and 3 more authors

Journal on Computing and Cultural Heritage, Apr 2021

Abs DOI PDF

This article reports on the experience of co-designing an educational video game aimed at promoting good dietary habits in youngsters and fostering Sustainable Development Goals (SDGs), such as SDG 3 (Good Health and Well-Being), SDG 10 (Reduced Inequalities), and SDG 17 (Partnerships for the Goals). To ensure the quality of the results, we developed a methodology under a social innovation paradigm that enabled the co-creation of the game. The methodology was driven by a series of three workshops, during which we adopted several different gamification strategies to support a Participatory Design (PD) process with the stakeholders, a group of local pre-teen and teen girls at social risk (N = 22). Captured requirements materialized into intermediate prototype evaluations that motivated a progressive refinement of the game.
A Project Review under the Focus of ‘Complexities’ on the Example of exploreAT!

Amelie Dorn, Eveline Wandl-Vogt, Thomas Palfinger, and 4 more authors

Digital Scholarship in the Humanities, Oct 2021

Abs DOI PDF

This article provides insights into dealing with complexities in the Digital Humanities project exploreAT!. By exploring a non-standard language collection for cultural insights, a three-fold approach is presented looking into concrete realizations and solutions of tackling challenges in terms of Open Innovation infrastructure, technology and the topic of choice, food. Methods and processes applied and developed in the project are aimed to serve as examples for future projects with similar data sets.
Analyzing and Visualizing Uncertain Knowledge: The Use of TEI Annotations in the PROVIDEDH Open Science Platform

Michał Kozak, Alejandro Rodríguez, Alejandro Benito-Santos, and 6 more authors

Journal of the Text Encoding Initiative, Mar 2021

Abs DOI PDF

The underlying uncertainty in digital humanities research data affects decision-making and persists during a project’s lifecycle. This uncertainty is inevitable since most empirical claims cannot be assessed against an absolute truth (Drucker 2011; Binder et al. 2014). This situation has been previously recognized together with the need to report the degrees of uncertainty that accompany such claims (Blau 2011). Although TEI makes it possible to annotate text with notions of certainty or precision, examples of actual projects taking advantage of this are scarce. There are many possible explanations for uncertainty’s lack of visibility in computationally supported humanities research; among them, the need for tools specifically designed to address the goal of defining and managing uncertainty stands out. Thus, efforts to provide technical support for humanities research should focus on managing and making uncertainty more transparent, rather than removing it. Another challenge is the fact that there is no agreement on a generic taxonomy for the different types of uncertainty that researchers may face. Various researchers across disciplines, working on varying projects and data sets, can use different categories to classify the uncertainties present in a particular case. In this paper, we introduce a collaborative platform for collective annotation of TEI data sets. We briefly present the flexible taxonomy of uncertainty used in the platform and describe two data sets used for its testing. Then we describe use cases of annotations available on the platform, and how they translate into TEI annotations. Creating and interpreting annotations with and without uncertainty should now be easier, especially for researchers who do not know TEI markup.

2020

A Data-Driven Introduction to Authors, Readings, and Techniques in Visualization for the Digital Humanities

A. Benito-Santos and Roberto Therón Sánchez

IEEE Computer Graphics and Applications, May 2020

Abs DOI PDF

The newly rediscovered frontier between data visualization and the digital humanities has proven to be an exciting field of experimentation for scholars from both disciplines. This fruitful collaboration is attracting researchers from other areas of science who may be willing to create visual analysis tools that promote humanities research in its many forms. However, as the collaboration grows in complexity, it may become intimidating for these scholars to get engaged in the discipline. To facilitate this task, we have built an introduction to visualization for the digital humanities that sits on a data-driven stance adopted by the authors. In order to construct a dataset representative of the discipline, we analyze citations from a core corpus on 300 publications in visualization for the humanities obtained from recent editions of the InfoVis Vis4DH workshop, the ADHO Digital Humanities Conference, and the specialized digital humanities journal Digital Humanities Quarterly. From here, we extract referenced works and analyze more than 1900 publications in search of citation patterns, prominent authors in the field, and other interesting insights. Finally, following the path set by other researchers in the visualization and Human-Computer Interaction (HCI) communities, we analyze paper keywords to identify significant themes and research opportunities in the field.
Defragmenting Research Areas with Knowledge Visualization and Visual Text Analytics

Alejandro Benito-Santos and Roberto Therón Sánchez

Applied Sciences, Jan 2020

Abs DOI PDF

The increasing specialization of science is motivating the fragmentation of traditional and well-established research areas into interdisciplinary communities of practice that focus on cooperation between experts to solve problems in a wide range of domains. This is the case of problem-driven visualization research (PDVR), in which groups of scholars use visualization techniques in different application domains such as the digital humanities, bioinformatics, sports science, or computer security. In this paper, we employ the findings obtained during the development of a novel visual text analytics tool we built in previous studies, GlassViz, to automatically detect interesting knowledge associations and groups of common interests between these communities of practice. Our proposed method relies on the statistical modeling of author-assigned keywords to make its findings, which are demonstrated in two use cases. The results show that it is possible to propose interactive, semisupervised visual approaches that aim at defragmenting a body of research using text-based, automatic literature analysis methods.
GlassViz: Visualizing Automatically-Extracted Entry Points for Exploring Scientific Corpora in Problem-Driven Visualization Research

A. Benito-Santos and R. Therón

In Proceedings of the 2020 IEEE Visualization Conference (VIS), Oct 2020

Abs DOI PDF

In this paper, we report the development of a model and a proof-of-concept visual text analytics (VTA) tool to enhance document discovery in a problem-driven visualization research (PDVR) context. The proposed model captures the cognitive model followed by domain and visualization experts by analyzing the interdisciplinary communication channel as represented by keywords found in two disjoint collections of research papers. High distributional intercollection similarities are employed to build informative keyword associations that serve as entry points to drive the exploration of a large document corpus. Our approach is demonstrated in the context of research on visualization for the digital humanities.
An Interactive Visualization of the Historical Dictionary of Bavarian Dialects in Austria

A. Benito-Santos, Antonio G. Losada Gómez, Roberto Theron Sánchez, and 2 more authors

In LingVis: Visual Analytics for Linguistics, 2020

PDF
Pilaster: A Collection of Citation Metadata Extracted From Publications on Visualization for the Digital Humanities

A. Benito-Santos and R. Therón

In Proceedings of the 2020 IEEE 5th Workshop on Visualization for the Digital Humanities (VIS4DH), Oct 2020

Abs DOI PDF

In this paper, we present Pilaster (https://visusal.github.io/pilaster/), a collection of citation metadata extracted from publications in visualization for the digital humanities. The collection is generated from a seed set of relevant publications from which we extracted cited works, including journal and conference papers, books, theses, or blog posts, among other resources. The main aim of this work revolves around three main points: first, the collection may serve as an entry point to the discipline for digital humanists and visualization scholars without previous experience in the field. Second, Pilaster can be regarded as a meeting point for more established visualization or humanities scholars seeking to collaborate in the development of novel research ideas and related visualization design studies in the context of the humanities. Third, and given the large amount of visualization design spaces that were captured, we believe the dataset has the potential to become the starting point for future studies aimed at understanding the particularities of problem-driven visualization research in this and other contexts.

2019

Evaluación de Una Metodología Novel Basada En Cuadernos Interactivos Para El Aprendizaje de La Programación En Python

Alejandro Benito, Rodrigo Santamaría, and Roberto Therón

Actas de las Jornadas sobre Enseñanza Universitaria de la Informática, 2019

Abs PDF

La enseñanza de la programación a alumnos de nivel preuniversitario sin experiencia en pensamiento computacional supone un reto que necesita de métodos didácticos noveles especialmente concebidos para un público que en los últimos años ha modificado radicalmente la forma en la que recibe información. Esta contribución describe la experiencia de adaptar y evaluar, en el contexto de una clase de alumnos de primer curso del Grado en Estadística (N=43), una metodología y un ecosistema de herramientas interactivas orientadas al aprendizaje de la programación y la ciencia de datos. Bajo este enfoque, la metodología experimental concebida implementa técnicas como la programación por pares y TDD (Test-Driven Development), lo cual, como se pudo comprobar a través de las evaluaciones realizadas, tuvo un efecto positivo en el aprovechamiento del tiempo dedicado a la asignatura y en la satisfacción general percibida por el alumnado sobre la impartición de los contenidos de la guía docente. En línea con otras investigaciones actuales, los materiales didácticos fueron puestos a disposición de la comunidad docente de manera abierta y gratuita, lo cual se espera contribuya a mejorar la metodología en futuras experiencias.

Abstract

Teaching programming to undergrad students without previous experience in computational thinking poses a challenge that calls for novel didactic methods oriented explicitly towards this kind of audience. In our contribution, we describe the experience of adapting and evaluating a novel methodology and its associated ecosystem of tools in a classroom of 43 first-year students (N=43) of statistics. Following recommendations found in the literature, the experimental methodology implements pair-programming and TDD (Test-

Driven Development) techniques intending to improve both the teaching and learning experiences. From the evaluation, we could learn that this had a positive impact on the students’ performance and their general satisfaction with the course. Finally, the educational materials of this research were collected and put online under a Creative Commons license so that other scientific peers can replicate our work in other future didactic experiences.
Visualización Automática de Estructuras Interdisciplinarias de Conocimiento a Través Del Análisis de La Semántica Latente de Las Palabras Clave

Alejandro Benito Santos

In Actas de La Jornada de Seguimiento de La Actividad Investigadora, May 2019

PDF
Cross-Domain Visual Exploration of Academic Corpora via the Latent Meaning of User-Authored Keywords

Alejandro Benito-Santos and Roberto Therón Sánchez

IEEE Access, 2019

Abs DOI PDF

Nowadays, scholars dedicate a substantial amount of their work to the querying and browsing of increasingly large collections of research papers on the Internet. In parallel, the recent surge of novel interdisciplinary approaches in science requires scholars to acquire competencies in new fields for which they may lack the necessary vocabulary to formulate adequate queries. This problem, together with the issue of information overload, poses new challenges in the fields of natural language processing (NLP) and visualization design that call for a rapid response from the scientific community. In this respect, we report on a novel visualization scheme that enables the exploration of research paper collections via the analysis of semantic proximity relationships found in author-assigned keywords. Our proposal replaces traditional string queries with a bag-of-words (BoW) extracted from a user-generated auxiliary corpus that captures the intentionality of the research. Continuing along the lines established by other authors in the fields of literature-based discovery (LBD), NLP, and visual analytics (VA), we combine novel advances in the fields of NLP with visual network analysis techniques to offer scholars a perspective of the target corpus that better fits their research interests. To highlight the advantages of our proposal, we conduct two experiments employing a collection of visualization research papers and an auxiliary cross-domain BoW. Here, we showcase how our visualization can be used to maximize the effectiveness of a browsing session by enhancing the language acquisition task, which allows for effectively extracting knowledge that is in line with the users’ previous expectations.
Exposing Uncertainty on the Historical Name Normalization Task

Alejandro Benito-Santos, Alejandro Rodríguez Díaz, and Roberto Therón Sánchez

In Proceedings of the Seventh International Conference on Technological Ecosystems for Enhancing Multiculturality, Oct 2019

Abs DOI PDF

As current research shows, the humanities and visualization communities are greatly benefitting from each other thanks to the interdisciplinary idiosyncrasy of a research field, the Digital Humanities (DH) that has gained significant attention in recent years from a vast number of scholars with disparate backgrounds. The DH suppose a vibrant field of experimentation for the application of innovative visualization design techniques and other related computational methods. A new wave in visualization research, uncertainty visualization, focuses on the display and conveyance of the uncertainty present in a great variety of statistical and algorithmic methods driving computations in a wide range of scientific domains. In this paper, we present a visualization dashboard concept in which we introduced uncertainty induced by a simple linguistic algorithm that is commonly employed in humanities research. According to this new paradigm, we showcase how accessible pre-existing visualizations, such as 2D histograms, can be easily adapted to communicate inherent algorithmic variability to the user, motivating uncertainty-aware research analyses in a DH context.
Intuitive Ontology-Based SPARQL Queries for RDF Data Exploration

Alejandro Rodríguez Díaz, Alejandro Benito-Santos, Amelie Dorn, and 3 more authors

IEEE Access, 2019

Abs DOI PDF

This paper introduces a strategy for both the retrieval and analysis of linked open data (LOD) based on the use of visual tools. Retrieving and understanding data from triplestores (such as SPARQL) requires technical knowledge and proves to be challenging with large datasets, which result in an increase in the mental overload when unknown ontologies are involved in the creation of complex queries. These two problems benefit greatly from visual techniques that allow for executing them in an easier and more intuitive manner. These techniques have already been applied to each problem separately; however, we propose combining them to lower the complexity of triple-store data retrieval and empower its exploration and analysis through specific data visualizations. To demonstrate the suitability of this strategy, a web-based tool was implemented. It allows for the creation of interactive queries using node-link ontology representations, and a subsequent filtering and analysis through a configurable dashboard with different data visualizations. This paper also presents a user-centred design process and evaluation of the proposed tool.

2018

Data-Driven Visual Performance Analysis in Soccer: An Exploratory Prototype

Alejandro Benito-Santos, Roberto Theron, Antonio Losada, and 2 more authors

Frontiers in Psychology, 2018

Abs DOI PDF

In soccer, understanding of collective tactical behaviour has become an integral part in sports analysis at elite levels. Evolution of technology allows collection of in-creasingly larger and more specific data sets related to sport activities in cost-effective and accessible manner. All this information is minutely scrutinized by thousands of analysts around the globe in search of an-swers that can in the long term help increase the performance of indi-viduals or teams in their respective competitions. As the volume of data increases in size, so does the complexity of the problem and the need for suitable tools that leverage the cognitive load involved in the inves-tigation. It is proven that visualization and computer-vision techniques, correctly applied to the context of a problem, help data analysts focus on the relevant information at each stage of the process, and generally lead to a better understanding of the facts that lie behind the data. In the current study, we presented a software prototype capable of assisting researchers and performance analysts in their duty of studying group collective behavior in soccer games and trainings. We used geospatial data acquired from a professional match to demonstrate its capabilities in two different case studies. Furthermore, we successfully proved the efficiency of the different visualization techniques implemented in the prototype and demonstrated how visual analysis can effectively improve some of the basic tasks employed by sports experts on their daily work, complementing more traditional approaches.
Toward Supporting Decision-making Under Uncertainty in Digital Humanities with Progressive Visualization

Roberto Therón, Alejandro Benito, Rodrigo Santamaría, and 1 more author

In Proceedings of the Sixth International Conference on Technological Ecosystems for Enhancing Multiculturality, 2018

Abs DOI PDF

Digital Humanities (DH) research and practice is subject to uncertainty during the life cycle of any project. Even in non data-oriented cases, analysts and other stakeholders need to make decisions without being aware of the level of uncertainty associated to the data being transformed by the computational tools used to enable the kind of novel work of humanists pursued within DH. We examine in this paper the literature that have characterized the types and sources of uncertainty in other fields, with the intent of establishing a foundation upon which build novel computational tools supporting the decision-making under uncertainty processes that DH is currently facing. We propose the use of progressive visual analytics as a feasible means to manage decision-making under uncertainty, which may help tackling some challenges related to the elimination or mitigation of uncertainty in DH, that otherwise would tamper the quality of the yielded results.

2017

Exploring Lemma Interconnections in Historical Dictionaries

Alejandro Benito, Roberto Therón, Antonio Losada, and 2 more authors

In Proc. 2nd Workshop on Visualization for the Digital Humanities (VIS4DH), 2017

Abs PDF

In this paper we discuss the goals, motivations and other particularities of a visual exploratory analysis tool for historical dictionaries of the Bavarian dialects in Austria. As an input data set we employ the digitized version of the Historical Dictionary of Bavarian Dialects in Austria (Wörterbuch der bairischen Mundarten in Österreich or WBO), an initiative started in 1963 which compiles more than five million paper slips collected during the years 1911-1998 in different areas of current Austria, the Czech Republic, Hungary and northern Italy. In the 1990s these paper slips started to be progressively digitized, becoming part of the Database of Bavarian Dialects (DBO) and in 2010 nearly 32.000 records related to plant names were made available online on the DBO electronically-mapped (dbo@ema) platform. In more recent efforts, the project exploreAT!: Exploring Austria’s Culture Through the Language Glass, which this work is part of, started in 2015. Its aim is “to reveal unique insights into the rich texture of the German Language, especially in Austria, by providing state of the art tools for exploring the unique collection (1911-1998) of the Bavarian Dialects in the region of the Austro-Hungarian Empire”, and it has originated other publications in conference proceedings that we encourage the reader to review [8, 19].
Co-Designing Innovation Networks for Cross-Sectoral Collaboration on the Example of exploreAT!

Caitlin Gura, Amelie Dorn, Alejandro Benito, and 2 more authors

In Proceedings of the 5th International Conference on Technological Ecosystems for Enhancing Multiculturality, Oct 2017

Abs DOI PDF

Daily interaction with digital media and devices has become a natural habit for the majority of people across different age ranges in Western cultures. The way we interact with and react to digital media, however, has been constantly evolving and there is a clear trend towards implementing device-oriented interactive engagement in a more meaningful way, where users readily add to shaping and creating digital experiences that not only generate added value for them personally, but also for the various other actors involved. This also concerns the interaction with both virtual as well as real-life objects in cultural settings. Across different sectors, creating a complementary synthesis of the digital and analog has been a primary topic, particularly in the course of the 21st century. In this paper we aim to discuss the establishment of cultural innovation networks between museums and other sectors, exemplify potential links between digital and physical objects in connection with a current Digital Humanities project (exploreAT!) and shed light on user engagement possibilities opening towards open innovation research infrastructures (OI-RI).

2016

A Spatio-Temporal Visual Analysis Tool for Historical Dictionaries

Alejandro Benito, Antonio G. Losada, Roberto Therón, and 3 more authors

In Proceedings of the Fourth International Conference on Technological Ecosystems for Enhancing Multiculturality, Nov 2016

Abs DOI PDF

The exploreAT! project aims to give insights into the richness of the German language in the Austrian area through a rich and unique collection of dialect words of the Bavarian dialects recorded during the former Austrian-Hungariann Monarchy period and beyond. Originally collected by means of questionnaires, words were noted in handwriting on individual paper slips, covering topics from nature and food to religious festivities, etc. Once digitized, the full database contains around 3.5 million single data entries with an estimated 200,000 headwords, which requires substantial effort if the analysts want to access specific information from the data set. It should also be noted that the data presents a high heterogeneity in terms of its nature and origin (from questionnaires, collectors, scientists, spoken language, hand written notes, etc.), which calls for the creation of a homogeneous database containing all of the available information. In this paper we present a tool aimed to improve the comprehension of that massive amount of data through visualization means, thus trying to help in the reach of meaningful conclusions and the acquisition of valuable insights in easy and fast ways. With it, analysts can discover cultural issues and access them through means of language and visualization. This is possible thanks to a multidimensional approach to data analysis based on the use of maps, projections and other visualization artifacts. To reach our goal, a team of experts with different backgrounds worked together trying to close the gap between the Humanities and Computer Sciences fields through the creation of our prototype and its multiple iterations.
BKViz: A Basketball Visual Analysis Tool

Antonio G. Losada, Roberto Therón, and Alejandro Benito

IEEE Computer Graphics and Applications, Nov 2016

Abs DOI PDF

The amount of data available in the sports field is difficult for coaches, analysts, and players to comprehend using classic analytics methods. Thus, new methods are necessary to help users break down that information and analyze it at a deeper level. The BKViz visual analytics system focuses on individual basketball games using classic and novel methods to reveal how players perform together and as individuals. The information is presented in interactive visualizations that allow immediate user feedback.