Adrian M.P. Brasoveanu

Background. Sentic computing relies on welldefined affective models of different complexity - polarity to distinguish positive and negative sentiment,
for example, or more nuanced models to capture expressions of human emotions. When used to measure
communication success, even the most granular affective model combined with sophisticated machine learning approaches may not fully capture an organisation’s
strategic positioning goals. Such goals often deviate from
the assumptions of standardised affective models. While
certain emotions such as Joy and Trust typically represent desirable brand associations, specific communication goals formulated by marketing professionals often
go beyond such standard dimensions. For instance, the
brand manager of a television show may consider fear
or sadness to be desired emotions for its audience.
Method. This article introduces expansion techniques
for affective models, combining common and commonsense knowledge available in knowledge graphs with
language models and affective reasoning, improving coverage and consistency as well as supporting domainspecific interpretations of emotions.
Results and Conclusions. An extensive evaluation
compares the performance of different expansion techniques: (i) a quantitative evaluation based on the revisited Hourglass of Emotions model to assess perfor
mance on complex models that cover multiple affective categories, using manually compiled gold standard
data, and (ii) a qualitative evaluation of a domainspecific affective model for television programme brands.
The results of these evaluations demonstrate that the
introduced techniques support a variety of embeddings
and pre-trained models. The paper concludes with a
discussion on applying this approach to other scenarios
where affective model resources are scarce.

Annotation styles express guidelines that direct human annotators in what rules to follow when creating gold standard annotations of text corpora. These guidelines not only shape the gold standards they help create, but also influence the training and evaluation of Named Entity Linking (NEL) tools, since different annotation styles correspond to divergent views on the entities present in the same texts. Such divergence is particularly present in texts from the media domain that contain references to creative works. In this work we present a corpus of 1000 annotated documents selected from the media domain. Each document is presented with multiple gold standard annotations representing various annotation styles. This corpus is used to evaluate a series of Named Entity Linking tools in order to understand the impact of the differences in annotation styles on the reported accuracy when processing highly ambiguous entities such as names of creative works. Relaxed annotation guidelines that include overlap styles lead to better results across all tools.

This paper explores the current landscape of visualizing large language models (LLMs). The main objective was threefold. Firstly, we investigate how we can visualize LLM-specific techniques such as prompt engineering, instruction tuning, or guidance. Secondly, LLM causality, interpretability, and explainability are examined through visualization. And finally, we showcase the role of visualization in illuminating the integration of multiple modalities. We are interested in discovering the papers that present visualization systems instead of those that use visualization to showcase a part of their work. Our survey aims to synthesize the state-of-the-art in LLM visualization, offering a compact resource for exploring future research avenues.

Visual techniques provide an intuitive way of making sense of the
large amounts of microposts available from social media sources,
particularly in the case of emerging topics of interest to a global
audience, which often raise controversy among key stakeholders.
Micropost streams are context-dependent and highly dynamic in nature. We describe a visual analytics platform to handle highvolume micropost streams from multiple social media channels. For each post we extract key contextual features such as location, topic and sentiment, and subsequently render the resulting multidimensional information space using a suite of coordinated views
that support a variety of complex information seeking behaviors. We also describe three new visualization techniques that extend the original platform to account for the dynamic nature of micropost streams through dynamic topography information landscapes, news flow diagrams and longitudinal cross-media analyses.

Most current evaluation tools are focused solely on benchmarking and comparative evaluations thus only provide aggregated statistics such as precision, recall and F1‐measure to assess overall system performance. They do not offer comprehensive analyses up to the level of individual annotations. This paper introduces Orbis, an extendable evaluation pipeline framework developed to allow visual drill‐down analyses of individual entities, computed by annotation services, in the context of the text they appear in, in reference to the entities specified in the gold standard.

The complexity of the socio-, political- and economical settings in which tourism enterprises operate, increasingly require them to make decisions that take into account data from various domains (e.g., economy, environmental sustainability). Based on a practitioners' survey that we performed, we conclude that although such cross-domain decisions are important, they are primarily performed by relying on manual data collection and aggregation, which is both time-consuming and error-prone. We propose a solution that relies on Linked Data as a technological platform for integrating data from three major tourism data sources: TourMIS, World Bank and Eurostat. Enabled by this integrated data, we developed the ETIHQ Dashboard, the first visual decision support system that supports cross-domain decisions over tourism, economic and sustainability indicators. An evaluation performed with practitioners shows that this Linked Data enabled systems brings important improvements in terms of execution times (28% faster) and answer quality when compared to current manual approaches.

In a global and interconnected economy, decision makers often need to consider information from various domains. A tourism destination manager, for example, has to correlate tourist behavior with financial and environmental indicators to allocate funds for strategic long-term investments. Statistical data underpins a broad range of such cross-domain decision tasks. A variety of statistical datasets are available as Linked Open Data, often incorporated into visual analytics solutions to support decision making. What are the principles, architectures, workflows and implementation design patterns that should be followed for building such visual cross-domain decision support systems. This article introduces a methodology to integrate and visualize cross-domain statistical data sources by applying selected RDF Data Cube (QB) principles. A visual dashboard built according to this methodology is presented and evaluated in the context of two use cases in the tourism and telecommunications domains.

The complexity of the social, political and economical settings in which tourism enterprises operate, increasingly require them to perform data analytics tasks that rely on data from various domains (e.g., economy, environmental sustainability). A survey of tourism practitioners performed in this study showed that although such cross-domain analytics are important, they are primarily performed by relying on manual data collection and aggregation, which is both time-consuming and error-prone. This paper investigates the suitability of Linked Data technologies to support data aggregation tasks needed for establishing such complex analytics systems. To that end, a prototypical implementation is developed that relies on Linked Data as a technological platform for integrating data from three major tourism data sources: TourMIS, World Bank and Eurostat. Enabled by this integrated data, the ETIHQ Dashboard for data analytics was implemented, the first visual data analytics system that supports cross-domain analytics over tourism, economic and sustainability indicators. An exploratory evaluation performed with practitioners shows that this Linked Data enabled system could potentially bring important improvements in terms of execution times and answer quality when compared to current manual approaches typically used by tourism practitioners in daily practice.

This paper introduces an interactive news flow visualization that reveals emerging topics in dynamic digital content archives. The presented approach combines several visual metaphors and can be easily adapted to present multi-source social media datasets. In the context of this work, we discuss various methods for improving visual interfaces for accessing aggregated media representations. We combine falling blocks with bar graphs and arcs, but keep these elements clearly separated in different areas of the display. The arc metaphor is adapted and enriched with interactive controls to help users understand the dataset's underlining meaning. The paper describes the implementation of the prototype and discusses design issues with a particular emphasis on visual metaphors to highlight hidden relations in digital content. We conclude with a summary of the lessons learnt and the integration of the visualization component into the Media Watch on Climate Change (www.ecoresearch.net/climate), a public Web portal that aggregates environmental information from a variety of online sources including news media, blogs and other social media such as Twitter, YouTube and Facebook.

Fake news detection is a difficult problem due to the nuances of language. Understanding the reasoning behind certain fake items implies inferring a lot of details about the various actors involved. We believe that the solution to this problem should be a hybrid one, combining machine learning, semantics and natural language processing. We introduce a new semantic fake news detection method built around relational features like sentiment, entities or facts extracted directly from text. Our experiments show that by adding semantic features the accuracy of fake news classification improves significantly.

The correct classification of the 17 Sustainable Development Goals (SDG) proposed by the United Nations (UN) is still a challenging and compelling prospect due to the Shared Task’s imbalanced dataset. This paper presents a good method to create a baseline using RoBERTa and data augmentation that offers a good overall performance on this imbalanced dataset. What is interesting to notice is that even though the alignment between synthetic gold and real gold was only marginally better than what would be expected by chance alone, the final scores were still okay.

Decision makers in the tourism domain routinely need to combine and compare statistical indicators about tourism and other related areas (e.g., economic). While many organizations offer relevant data sets, their automatic access and reuse is hampered (i) by them being offered as data dumps in non-semantic encodings; (ii) by them assuming some implicit knowledge that is necessary to build applications (e.g., that a city is situated in a certain country) and (iii) by the use of incompatible ways to measure the same indicator without formally specifying the assumptions behind the measurement technique. We explore the use of linked data technologies to solve these issues by triplifying the content of TourMIS, a broadly used data source of European tourism statistics and by building a prototype system using this data.

The TourMISLOD dataset exposes as linked data a significant portion of the content of TourMIS, a key source of European tourism statistics data. TourMISLOD contains information about the Arrivals, Bednights and Capacity tourism indicators, recorded from 1985 onwards, about over 150 European cities and in connection to 19 major markets. Due to licensing issues, the usage of this dataset is currently limited to the TourMIS consortium, however, a prototype application has already revealed its usefulness for decision support.

Identifying all names that refer to a particular set of named entities is a challenging task, as quite often we need to consider many features that include a lot of variation like abbreviations, aliases, hypocorism, multilingualism or partial matches. Each entity type can also have specific rules for
name variances: people names can include titles, country and branch names are sometimes removed from organization names, while locations are often plagued by the issue of nested entities. The lack of a clear strategy for collecting, processing and computing name variants significantly lowers the
recall of tasks such as Named Entity Linking and Knowledge Base Population since name variances are frequently used in all kind of textual content.
This paper proposes several strategies to address these issues. Recall can be improved by combining knowledge repositories and by computing additional variances based on algorithmic approaches. Heuristics and machine learning methods then analyze the generated name variances and mark ambiguous names to increase precision. An extensive evaluation demonstrates the effects
of integrating these methods into a new Named Entity Linking framework and confirms that systematically considering name variances yields significant performance improvements.

The news media landscape tends to focus on long-running narratives. Correctly processing new information, therefore, requires considering multiple lenses when analyzing media content. Traditionally it would have been considered sufficient to extract the topics or entities contained in a text in order to classify it, but today it is important to also look at more sophisticated annotations related to fine-grained geolocation, events, stories and the relations between them. In order to leverage such lenses we propose a new corpus that offers a diverse set of annotations over texts collected from multiple media sources. We also showcase the framework used for creating the corpus, as well as how the information from the various lenses can be used in order to support different use cases in the EU project InVID for verifying the veracity of online video.