Search

The nuances of languages, as well as the varying degrees of truth observed in news items, make fake news detection a difficult problem to solve. A news item is never launched without a purpose, therefore in order to understand its motivation it is best to analyze the relations between the speaker and its subject, as well as different credibility metrics. Inferring details about the various actors involved in a news item is a problem that requires a hybrid approach that mixes machine learning, semantics and natural language processing. This article discusses a semantic fake news detection method built around relational features like sentiment, entities or facts extracted directly from text. Our experiments are focused on short texts with different degrees of truth and show that adding semantic features improves accuracy significantly.

Automatic extraction of forum posts and metadata is a crucial but challenging task since forums do not expose their content in a standardized structure. Content extraction methods, therefore, often need customizations such as adaptations to page templates and improvements of their extraction code before they can be deployed to new forums. Most of the current solutions are also built for the more general case of content extraction from web pages and lack key features important for understanding forum content such as the identification of author metadata and information on the thread structure. This paper, therefore, presents a method that determines the XPath of forum posts, eliminating incorrect mergers and splits of the extracted posts that were common in systems from the previous generation. Based on the individual posts further metadata such as authors, forum URL and structure are extracted. We also introduce Harvest, a new open source toolkit that implements the presented methods and create a gold standard extracted from 52 different Web forums for evaluating our approach. A comprehensive evaluation reveals that Harvest clearly outperforms competing systems., IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT 2020), Accepted 27 October 2020

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.

Knowledge-rich Information Extraction (IE) methods aspire towards combining classical IE with background knowledge obtained from third-party resources. Linked Open Data repositories that encode billions of machine readable facts from sources such as Wikipedia play a pivotal role in this development. The recent growth of Linked Data adoption for Information Extraction tasks has shed light on many data quality issues in these data sources that seriously challenge their usefulness such as completeness, timeliness and semantic correctness. Information Extraction methods are, therefore, faced with problems such as name variance and type confusability. If multiple linked data sources are used in parallel, additional concerns regarding link stability and entity mappings emerge. This paper develops methods for integrating Linked Data into Named Entity Linking methods and addresses challenges in regard to mining knowledge from Linked Data, mitigating data quality issues, and adapting algorithms to leverage this knowledge. Finally, we apply these methods to Recognyze, a graph-based Named Entity Linking (NEL) system, and provide a comprehensive evaluation which compares its performance to other well-known NEL systems, demonstrating the impact of the suggested methods on its own entity linking performance.

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 topicsor 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.

Over the last years, the amount of data published as Linked Data on the Web has grown enormously. In spite of the high availability of Linked Data, organizations still encounter an accessibility challenge while consuming it. This is mostly due to the large size of some of the datasets published as Linked Data. The core observation behind this work is that a subset of these datasets suffices to address the needs of most organizations. In this paper, we introduce Torpedo, an approach for efficiently selecting and extracting relevant subsets from RDF datasets. In particular, Torpedo adds optimization techniques to reduce seek operations costs as well as the support of multi-join graph patterns and SPARQL FILTERs that enable to perform a more granular data selection. We compare the performance of our approach with existing solutions on nine different queries against four datasets. Our results show that our approach is highly scalable and is up to 26% faster than the current state-of-the-art RDF dataset slicing approach.

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.

Watching the evolution of the Semantic Web (SW) from its inception to these days we can easily observe that the main task the developers face while building it is to encode the human knowledge into ontologies and the human reasoning into dedicated reasoning engines. Now, the SW needs to have efficient mechanisms to access information by both humans and artificial agents. The most important tools in this context are ontologies. The last years have been dedicated to solving the infrastructure problems related to ontologies: ontology management, ontology matching, ontology adoption, but as time goes by and these problems are better understood the research interests in this area will surely shift towards the way in which agents will use them to communicate between them and with humans. Despite the fact that interface agents could be bilingual, it would be more efficient, safe and swift that they should use the same language to communicate with humans and with their peers. Since anthropocentric systems entail nowadays multimodal interfaces, it seems suitable to build multimodal ontologies. Generic ontologies are needed when dealing with uncertainty. Multimodal ontologies should be designed taking into account our way of thinking (mind maps, visual thinking, feedback, logic, emotions, etc.) and also the processes in which they would be involved (multimodal fusion and integration, error reduction, natural language processing, multimodal fission, etc.). By doing this it would be easier for us (and also fun) to use ontologies, but in the same time the communication with agents (and also agent to agent talk) would be enhanced. This is just one of our conclusions related to why building generic multimodal ontologies is very important for future semantic web applications.

Rigorous evaluations and analyses of evaluation results are key towards improving Named Entity Linking systems. Nevertheless, most current evaluation tools are focused on benchmarking and comparative evaluations. Therefore, they only provide aggregated statistics such as precision, recall and F1-measure to assess system performance and no means for conducting detailed analyses up to the level of individual annotations. This paper addresses the need for transparent benchmarking and fine-grained error analysis by introducing Orbis, an extensible framework that supports drill-down analysis, multiple annotation tasks and resource versioning. Orbis complements approaches like those deployed through the GERBIL and TAC KBP tools and helps developers to better understand and address shortcomings in their Named Entity Linking tools. We present three uses cases in order to demonstrate the usefulness of Orbis for both research and production systems: (i) improving Named Entity Linking tools; (ii) detecting gold standard errors; and (iii) performing Named Entity Linking evaluations with multiple versions of the included resources.

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.