Detecting salient parts in text using natural language processing has been widely used to mitigate the effects of information overflow. Nevertheless, most of the datasets available for this task are derived mainly from academic publications. We introduce SPACE-IDEAS, a dataset for salient information detection from innovation ideas related to the Space domain. The text in SPACE-IDEAS varies greatly and includes informal, technical, academic and business-oriented writing styles. In addition to a manually annotated dataset we release an extended version that is annotated using a large generative language model. We train different sentence and sequential sentence classifiers, and show that the automatically annotated dataset can be leveraged using multitask learning to train better classifiers.

Preventing the spread of misinformation is challenging. The detection of misleading content presents a significant hurdle due to its extreme linguistic and domain variability. Content-based models have managed to identify deceptive language by learning representations from textual data such as social media posts and web articles. However, aggregating representative samples of this heterogeneous phenomenon and implementing effective real-world applications is still elusive. Based on analytical work on the language of misinformation, this paper analyzes the linguistic attributes that characterize this phenomenon and how representative of such features some of the most popular misinformation datasets are. We demonstrate that the appropriate use of pertinent symbolic knowledge in combination with neural language models is helpful in detecting misleading content. Our results achieve state-of-the-art performance in misinformation datasets across the board, showing that our approach offers a valid and robust alternative to multi-task transfer learning without requiring any additional training data. Furthermore, our results show evidence that structured knowledge can provide the extra boost required to address a complex and unpredictable real-world problem like misinformation detection, not only in terms of accuracy but also time efficiency and resource utilization.

We present SpaceQA, to the best of our knowledge the first open-domain QA system in Space mission design. SpaceQA is part of an initiative by the European Space Agency (ESA) to facilitate the access, sharing and reuse of information about Space mission design within the agency and with the public. We adopt a state-of-the-art architecture consisting of a dense retriever and a neural reader and opt for an approach based on transfer learning rather than fine-tuning due to the lack of domain-specific annotated data. Our evaluation on a test set produced by ESA is largely consistent with the results originally reported by the evaluated retrievers and confirms the need of fine tuning for reading comprehension. As of writing this paper, ESA is piloting SpaceQA internally.

This paper summarises work where we combined semantic web technologies with deep learning systems to obtain state-of-the art explainable misinformation detection. We proposed a conceptual and computational model to describe a wide range of misinformation detection systems based around the concepts of credibility and reviews. We described how Credibility Reviews (CRs) can be used to build networks of distributed bots that collaborate for misinformation detection which we evaluated by building a prototype based on publicly available datasets and deep learning models.

Textbook Question Answering is a complex task in the intersection of Machine Comprehension and Visual Question Answering that requires reasoning with multimodal information from text and diagrams.For the first time, this paper taps on the potential of transformer language models and bottom-up and top-down attention to tackle the language and visual understanding challenges this task entails.Rather than training a language-visual transformer from scratch we rely on pretrained transformers, fine-tuning and ensembling.We add bottom-up and top-down attention to identify regions of interest corresponding to diagram constituents and their relationships, improving the selection of relevant visual information for each question and answer options.Our system ISAAQ reports unprecedented success in all TQA question types, with accuracies of 81.36%, 71.11% and 55.12% on true/false, text-only and diagram multiple choice questions.ISAAQ also demonstrates its broad applicability, obtaining state-of-theart results in other demanding datasets.

The current scientific and technological landscape is characterised by the increasing availability of data resources and processing tools and services. In this setting, metadata have emerged as a key factor facilitating management, sharing and usage of such digital assets. In this paper we present ELG-SHARE, a rich metadata schema catering for the description of Language Resources and Technologies (processing and generation services and tools, models, corpora, term lists, etc.), as well as related entities (e.g., organizations, projects, supporting documents, etc.). The schema powers the European Language Grid platform that aims to be the primary hub and marketplace for industry-relevant Language Technology in Europe. ELG-SHARE has been based on various metadata schemas, vocabularies, and ontologies, as well as related recommendations and guidelines.

With 24 official EU and many additional languages, multilingualism in Europe and an inclusive Digital Single Market can only be enabled through Language Technologies (LTs). European LT business is dominated by hundreds of SMEs and a few large players. Many are world-class, with technologies that outperform the global players. However, European LT business is also fragmented – by nation states, languages, verticals and sectors, significantly holding back its impact. The European Language Grid (ELG) project addresses this fragmentation by establishing the ELG as the primary platform for LT in Europe. The ELG is a scalable cloud platform, providing, in an easy-to-integrate way, access to hundreds of commercial and non-commercial LTs for all European languages, including running tools and services as well as data sets and resources. Once fully operational, it will enable the commercial and non-commercial European LT community to deposit and upload their technologies and data sets into the ELG, to deploy them through the grid, and to connect with other resources. The ELG will boost the Multilingual Digital Single Market towards a thriving European LT community, creating new jobs and opportunities. Furthermore, the ELG project organises two open calls for up to 20 pilot projects. It also sets up 32 national competence centres and the European LT Council for outreach and coordination purposes.

Multilingualism is a cultural cornerstone of Europe and firmly anchored in the European treaties including full language equality. However, language barriers impacting business, cross-lingual and cross-cultural communication are still omnipresent. Language Technologies (LTs) are a powerful means to break down these barriers. While the last decade has seen various initiatives that created a multitude of approaches and technologies tailored to Europe’s specific needs, there is still an immense level of fragmentation. At the same time, AI has become an increasingly important concept in the European Information and Communication Technology area. For a few years now, AI – including many opportunities, synergies but also misconceptions – has been overshadowing every other topic. We present an overview of the European LT landscape, describing funding programmes, activities, actions and challenges in the different countries with regard to LT, including the current state of play in industry and the LT market. We present a brief overview of the main LT-related activities on the EU level in the last ten years and develop strategic guidance with regard to four key dimensions.

Deep learning currently dominates the benchmarks for various NLP tasks and, at the basis of such systems, words are frequently represented as embeddings --vectors in a low dimensional space-- learned from large text corpora and various algorithms have been proposed to learn both word and concept embeddings. One of the claimed benefits of such embeddings is that they capture knowledge about semantic relations. Such embeddings are most often evaluated through tasks such as predicting human-rated similarity and analogy which only test a few, often ill-defined, relations. In this paper, we propose a method for (i) reliably generating word and concept pair datasets for a wide number of relations by using a knowledge graph and (ii) evaluating to what extent pre-trained embeddings capture those relations. We evaluate the approach against a proprietary and a public knowledge graph and analyze the results, showing which lexico-semantic relational knowledge is captured by current embedding learning approaches.

Compared to natural images, understanding scientific figures is particularly hard for machines. However, there is a valuable source of information in scientific literature that until now has remained untapped: the correspondence between a figure and its caption. In this paper we investigate what can be learnt by looking at a large number of figures and reading their captions, and introduce a figure-caption correspondence learning task that makes use of our observations. Training visual and language networks without supervision other than pairs of unconstrained figures and captions is shown to successfully solve this task. We also show that transferring lexical and semantic knowledge from a knowledge graph significantly enriches the resulting features. Finally, we demonstrate the positive impact of such features in other tasks involving scientific text and figures, like multi-modal classification and machine comprehension for question answering, outperforming supervised baselines and ad-hoc approaches.

Data-intensive science disciplines, like Earth Science, are increasingly producing and consuming a variety of digital resources during the course of a scientific investigation. Instead of having these resources in isolated repositories, scientists are seeking ways for managing and making these resources available from a single place, and at the same time they are also increasingly interested in the adoption of FAIR principles to enhance the visibility and reusability of scientific results. This has called for new methods to improve the access and communication of results. Research Objects are a key building block towards realising this vision. They provide a structured way (a model) to describe scientific resources related to an investigation, along with the context in which they were used and the people involved. But research objects are as useful in practice as the availability of tools supporting their adoption. In this paper, we present a toolkit, tailored for Earth Sciences, comprising a set of services and applications around research objects that support scientists throughout the research lifecycle to manage, share, find and reuse scientific results, and we discuss initial insights into the community adoption.

Research objects were designed in data-intensive science under the premises of interoperability and machine-readability to describe scientific processes and findings including all the resources that were used in the research endeavour. In this poster we present our work with Earth Science communities, that have embraced the research object model for long-term preservation and reuse of knowledge, to design checklists, the main tool available to assess the quality of research objects and upon which quality metrics such completeness, stability and reliability are calculated.

Research objects act as a semantically rich container of all the information that lead to a scientific result, and have the potential to change how data intensive science shares and reuses methods, datasets and results. Despite a comprehensive set of vocabularies to describe semantically the aggregated resources, users often limit themselves to provide metadata at the container level, ignoring the valuable information that the aggregated resources contain. In this poster we explore how the combination of semantic technologies and natural language processing can be used to enrich research objects with structured metadata aiming at enhancing their findability as a crucial aspect towards their reuse.

The provisioning and exploitation at a global scale of environmental and geophysical data requires advanced automation and governance mechanisms that enable (meta)data interoperability but also the exchange of formalized scientific concepts and methods. In this paper we introduce recent efforts in such direction, based on scientific workflows and research objects as enablers of such vision. The former enable the integration of different web services in a higher-level data processing artifact while the latter enhances governance around data product validation and consistency, result reproducibility and credit to the principal investigators and data providers. This paper provides a concise overview of our project, current status and next steps.

A research object is a single information unit encapsulating all the knowledge relevant to a particular scientific investigation, their associated metadata and the context where such resources were produced and came into play. Aimed at enhancing the preservation, reuse and scholarly communication of data-intensive science, research objects are both technical and social artifacts that represent a partnership between scientific communities and the computational support required in nowadays science. In this paper, we explore such partnership, identifying the lack of appropriate machine-readable metadata as one of its main inhibitors, and address the semantic enrichment of research objects as one key aspect towards its establishment. Focused on the specific needs of Earth Science communities, we propose extensions to research object representation models and present novel methods and tools to enrich research object metadata through automatic means. Finally, we validate the approach through the implementation of a recommender system that exploits the resulting metadata to facilitate research object discovery and reuse, enabling humans and machines to work together and accelerate the research life cycle.

Earth science disciplines are increasingly producing large amounts of heterogeneous data that needs to be accessed, integrated and processed during the course of a research or operational process. In this setting, there is a growing need from earth scientists to collect and manage effectively these resources, including the data used, the methods applied, and the results produced. This is even more challenging if we consider i) the dynamic and collaborative nature of the earth science processes; ii) the interfaces typically used by scientists where they should be able to accomplish core management tasks with a minimal overhead. In this poster, we present our approach to address these challenges, which is based on research objects and on the associated technological support provided by ROHub platform. We show how ROHub is used as the underlying technology for supporting earth scientists work through different interfaces tailored to their communities needs and usage scenario.

Scientific workflows are a popular mechanism for specifying and automating data-driven in silico experiments. A significant aspect of their value lies in their potential to be reused. Once shared, workflows become useful building blocks that can be combined or modified for developing new experiments. However, previous studies have shown that storing workflow specifications alone is not sufficient to ensure that they can be successfully reused, without being able to understand what the workflows aim to achieve or to re-enact them. To gain an understanding of the workflow, and how it may be used and repurposed for their needs, scientists require access to additional resources such as annotations describing the workflow, datasets used and produced by the workflow, and provenance traces recording workflow executions. In this article, we present a novel approach to the preservation of scientific workflows through the application of research objects—aggregations of data and metadata that enrich the workflow specifications. Our approach is realised as a suite of ontologies that support the creation of workflow-centric research objects. Their design was guided by requirements elicited from previous empirical analyses of workflow decay and repair. The ontologies developed make use of and extend existing well known ontologies, namely the Object Reuse and Exchange (ORE) vocabulary, the Annotation Ontology (AO) and the W3C PROV ontology (PROVO). We illustrate the application of the ontologies for building Workflow Research Objects with a case-study that investigates Huntington’s disease, performed in collaboration with a team from the Leiden University Medial Centre (HG-LUMC). Finally we present a number of tools developed for creating and managing workflow-centric research objects.

In this paper we present a novel semi-automatic framework for acquiring and maintaining fuzzy ontological knowledge in dialogue systems. The main goal is to narrow the vagueness interpretation gap between the systems and their users and thus offer better information services to the latter.

Enabling Subject Matter Experts (SMEs) to formulate knowledge without the intervention of Knowledge Engineers (KEs) requires providing SMEs with methods and tools that abstract the underlying knowledge representation and allow them to focus on modeling activities. Bridging the gap between SME-authored models and their representation is challenging, especially in the case of complex knowledge types like processes, where aspects like frame management, data, and control flow need to be addressed. In this paper, we describe how SME-authored process models can be provided with an operational semantics and grounded in a knowledge representation language like F-logic to support process-related reasoning. The main results of this work include a formalism for process representation and a mechanism for automatically translating process diagrams into executable code following such formalism. From all the process models authored by SMEs during evaluation 82 percent were well formed, all of which executed correctly. Additionally, the two optimizations applied to the code generation mechanism produced a performance improvement at reasoning time of 25 and 30 percent with respect to the base case, respectively.

Workflows provide a popular means for preserving scientific methods by explicitly encoding their process. However, some of them are subject to a decay in their ability to be re-executed or reproduce the same results over time, largely due to the volatility of the resources required for workflow executions. This paper provides an analysis of the root causes of workflow decay based on an empirical study of a collection of Taverna workflows from the myExperiment repository. Although our analysis was based on a specific type of workflow, the outcomes and methodology should be applicable to workflows from other systems, at least those whose executions also rely largely on accessing third-party resources. Based on our understanding about decay we recommend a minimal set of auxiliary resources to be preserved together with the workflows as an aggregation object and provide a software tool for end-users to create such aggregations and to assess their completeness.

In this paper we present an approach towards knowledgeacquisition of process knowledge for the natural sciences.The work has been conducted within Project Halo, whichis creating advanced knowledge authoring and questionanswering systems for the natural sciences. An analysis of AP®-level questions for Biology, Chemistry and Physicsuncovered that process knowledge is the single most frequenttype of knowledge required. Thus, we developedmeans to acquire process knowledge, to formally representit, and to reason about it in order to answer novel questionsabout the domains.All these tasks are supported by an abstract process metamodel.It provides the terminology for user-tailored processdiagrams, which are automatically translated into executableFLogic code. The meta-model and the code generationare based on the notion of Problem Solving Methods(PSM) which represent an abstract formalization of thereasoning strategies needed for processes.