Ontologies of research topics are crucial for structuring scientific knowledge, enabling scientists to navigate vast amounts of research, and forming the backbone of intelligent systems such as search engines and recommendation systems. However, manual creation of these ontologies is expensive, slow, and often results in outdated and overly general representations. As a solution, researchers have been investigating ways to automate or semi-automate the process of generating these ontologies. One of the key challenges in this domain is accurately assessing the semantic relationships between pairs of research topics. This paper presents an analysis of the capabilities of large language models (LLMs) in identifying such relationships, with a specific focus on the field of engineering. To this end, we introduce a novel benchmark based on the IEEE Thesaurus for evaluating the task of identifying three types of semantic relations between pairs of topics: broader , narrower , and same-as . Our study evaluates the performance of seventeen LLMs, which differ in scale, accessibility (open vs. proprietary), and model type (full vs. quantised), while also assessing four zero-shot reasoning strategies. Several models with varying architectures and sizes have achieved excellent results on this task, including Mixtral-8 × 7B, Dolphin-Mistral-7B, and Claude 3 Sonnet, with F1-scores of 0.847, 0.920, and 0.967, respectively. Furthermore, our findings demonstrate that smaller, quantised models, when optimised through prompt engineering, can achieve strong performance while requiring very limited computational resources.

In scientific papers, it is common practice to cite other articles to substantiate claims, provide evidence for factual assertions, reference limitations, and research gaps, and fulfill various other purposes. When authors include a citation in a given sentence, there are two considerations they need to take into account: (i) where in the sentence to place the citation and (ii) which citation to choose to support the underlying claim. In this paper, we focus on the first task as it allows multiple potential approaches that rely on the researcher’s individual style and the specific norms and conventions of the relevant scientific community. We propose two automatic methodologies that leverage transformers architecture for either solving a Mask-Filling problem or a Named Entity Recognition problem. On top of the results of the proposed methodologies, we apply ad-hoc Natural Language Processing heuristics to further improve their outcome. We also introduce s2orc-9K, an open dataset for fine-tuning models on this task. A formal evaluation demonstrates that the generative approach significantly outperforms five alternative methods when fine-tuned on the novel dataset. Furthermore, this model’s results show no statistically significant deviation from the outputs of three senior researchers.

The design and management of modern big data platforms are extremely complex. It requires carefully integrating multiple storage and computational platforms as well as implementing approaches to protect and audit data access. Therefore, onboarding new data and implementing new data transformation processes is typically time-consuming and expensive. In many cases, enterprises construct their data platforms without a clear distinction between logical and technical concerns. Consequently, these platforms lack sufficient abstraction and are closely tied to particular technologies, making the adaptation to technological evolution very costly. This paper illustrates a novel approach to designing data platform models based on a formal ontology that structures various domain components into an accessible knowledge graph. We also describe the preliminary version of AGILE-DM, a novel ontology that we built for this purpose. Our solution is flexible, technologically agnostic, and more adaptable to changes and technical advancements.

Quality journalism has become more important than ever due to the need for quality and trustworthy media outlets that can provide accurate information to the public and help to address and counterbalance the wide and rapid spread of disinformation. At the same time, quality journalism is under pressure due to loss of revenue and competition from alternative information providers. This vision paper discusses how recent advances in Artificial Intelligence (AI), and in Machine Learning (ML) in particular, can be harnessed to support efficient production of high-quality journalism. From a news consumer perspective, the key parameter here concerns the degree of trust that is engendered by quality news production. For this reason, the paper will discuss how AI techniques can be applied to all aspects of news, at all stages of its production cycle, to increase trust.

During an emergency event, such as a fire evacuation, support services benefit from having information about people who may require special assistance. In this context, health data represents a particularly important source of information, as it can allow an emergency response system to build an accurate picture of people's relevant health conditions and use this to advise responders. However, to perform this task, a system needs to represent and reason over the evolution of health conditions over time. Crucially, it needs to predict the probability that a potentially relevant condition mentioned in a health record is still valid at the time of the emergency. In this paper, we propose a methodology for representing the evolution of health conditions and reasoning about them in the context of an emergency scenario. To support our approach with data, we develop a pipeline to capture knowledge about condition evolution from reliable sources in natural language. We incorporate these two components into a system that predicts a person's likelihood of being vulnerable during an emergency event. Finally, we demonstrate that representing and reasoning about condition evolution improves the quality and precision of the recommendations provided by our system to emergency services.

Scientific knowledge has been traditionally disseminated and preserved through research articles published in journals, conference proceedings, and online archives. However, this article-centric paradigm has been often criticized for not allowing to automatically process, categorize, and reason on this knowledge. An alternative vision is to generate a semantically rich and interlinked description of the content of research publications. In this paper, we present the Artificial Intelligence Knowledge Graph (AI-KG), a large-scale automatically generated knowledge graph that describes 820K research entities. AI-KG includes about 14M RDF triples and 1.2M reified statements extracted from 333K research publications in the field of AI, and describes 5 types of entities (tasks, methods, metrics, materials, others) linked by 27 relations. AI-KG has been designed to support a variety of intelligent services for analyzing and making sense of research dynamics, supporting researchers in their daily job, and helping to inform decision-making in funding bodies and research policymakers. AI-KG has been generated by applying an automatic pipeline that extracts entities and relationships using three tools: DyGIE++, Stanford CoreNLP, and the CSO Classifier. It then integrates and filters the resulting triples using a combination of deep learning and semantic technologies in order to produce a high-quality knowledge graph. This pipeline was evaluated on a manually crafted gold standard, yielding competitive results. AI-KG is available under CC BY 4.0 and can be downloaded as a dump or queried via a SPARQL endpoint.

The task of identifying pieces of evidence in texts is of fundamental importance in supporting qualitative studies in various domains, especially in the humanities. In this paper, we coin the expression themed evidence, to refer to (direct or indirect) traces of a fact or situation relevant to a theme of interest and study the problem of identifying them in texts. We devise a generic framework aimed at capturing themed evidence in texts based on a hybrid approach, combining statistical natural language processing, background knowledge, and Semantic Web technologies. The effectiveness of the method is demonstrated on a case study of a digital humanities database aimed at collecting and curating a repository of evidence of experiences of listening to music. Extensive experiments demonstrate that our hybrid approach outperforms alternative solutions. We also evidence its generality by testing it on a different use case in the digital humanities.

One of the existing query recommendation strategies for unknown datasets is "by example", i.e. based on a query that the user already knows how to formulate on another dataset within a similar domain. In this paper we measure what contribution a structural analysis of the query and the datasets can bring to a recommendation strategy, to go alongside approaches that provide a semantic analysis. Here we concentrate on the case of star-shaped SPARQL queries over RDF datasets. The illustrated strategy performs a least general generalization on the given query, computes the specializations of it that are satisfiable by the target dataset, and organizes them into a graph. It then visits the graph to recommend first the reformulated queries that reflect the original query as closely as possible. This approach does not rely upon a semantic mapping between the two datasets. An implementation as part of the SQUIRE query recommendation library is discussed.

Ontologies of research areas are important tools for characterising, exploring, and analysing the research landscape. Some fields of research are comprehensively described by large-scale taxonomies, e.g., MeSH in Biology and PhySH in Physics. Conversely, current Computer Science taxonomies are coarse-grained and tend to evolve slowly. For instance, the ACM classification scheme contains only about 2K research topics and the last version dates back to 2012. In this paper, we introduce the Computer Science Ontology (CSO), a large-scale, automatically generated ontology of research areas, which includes about 26K topics and 226K semantic relationships. It was created by applying the Klink-2 algorithm on a very large dataset of 16M scientific articles. CSO presents two main advantages over the alternatives: (i) it includes a very large number of topics that do not appear in other classifications, and (ii) it can be updated automatically by running Klink-2 on recent corpora of publications. CSO powers several tools adopted by the editorial team at Springer Nature and has been used to enable a variety of solutions, such as classifying research publications, detecting research communities, and predicting research trends. To facilitate the uptake of CSO we have developed the CSO Portal, a web application that enables users to download, explore, and provide granular feedback on CSO at different levels. Users can use the portal to rate topics and relationships, suggest missing relationships, and visualise sections of the ontology. The portal will support the publication of and access to regular new releases of CSO, with the aim of providing a comprehensive resource to the various communities engaged with scholarly data.

When publishing data, data licences are used to specify the actions that are permitted or prohibited, and the duties that target data consumers must comply with. However, in complex environments such as a smart city data portal, multiple data sources are constantly being combined, processed and redistributed. In such a scenario, deciding which policies apply to the output of a process based on the licences attached to its input data is a difficult, knowledge-intensive task. In this paper, we evaluate how automatic reasoning upon semantic representations of policies and of data flows could support decision making on policy propagation. We report on the results of a user study designed to assess both the accuracy and the utility of such a policy-propagation tool, in comparison to a manual approach.

Technologies such as algorithms, applications and formats are an important part of the knowledge produced and reused in the research process. Typically, a technology is expected to originate in the context of a research area and then spread and contribute to several other fields. For example, Semantic Web technologies have been successfully adopted by a variety of fields, e.g., Information Retrieval, Human Computer Interaction, Biology, and many others. Unfortunately, the spreading of technologies across research areas may be a slow and inefficient process, since it is easy for researchers to be unaware of potentially relevant solutions produced by other research communities. In this paper, we hypothesise that it is possible to learn typical technology propagation patterns from historical data and to exploit this knowledge i) to anticipate where a technology may be adopted next and ii) to alert relevant stakeholders about emerging and relevant technologies in other fields. To do so, we propose the Technology-Topic Framework, a novel approach which uses a semantically enhanced technology-topic model to forecast the propagation of technologies to research areas. A formal evaluation of the approach on a set of technologies in the Semantic Web and Artificial Intelligence areas has produced excellent results, confirming the validity of our solution.

The focus of this work is to exploit ontologies to make robotic systems more accessible to non-expert users, therefore supporting the deployment of robot-integrated applications. Due to the increasing number of robotic platforms available for commercial use, robotic systems are nowadays being approached by users with different backgrounds, who are often more interested in the robots' high-level capabilities than their technical architecture. Without the right expertise however, using robots is restricted to the capabilities exposed by the platform provider, i.e. they can only be used as end products rather than as development platforms. Our hypothesis is that an ontological representation of the capabilities of robots could make these capabilities more accessible, reducing the complexity of robot programming and enabling non-experts to exploit these systems to a much larger extent. To demonstrate this, an ontology abstracting the capabilities exposed by the most common robotic middleware (ROS) is integrated in a system to allow non-experts to program robots of different types and capabilities without previous knowledge either of the specific robotic platform being considered, or of the intricate systems used in its implementation. Our experiments, in which non-experts users had to configure the system in order to make robots achieve different tasks, show how the efforts required for realizing basic tasks using available robotic platforms can be sensibly reduced through our approach.

Governing the life cycle of data on the web is a challenging issue for organisations and users. Data is distributed under certain policies that determine what actions are allowed and in which circumstances. Assessing what policies propagate to the output of a process is one crucial problem. Having a description of policies and data flow steps implies a huge number of propagation rules to be specified and computed (number of policies times number of actions). In this paper we provide a method to obtain an abstraction that allows to reduce the number of rules significantly. We use the Datanode ontology, a hierarchical organisation of the possible relations between data objects, to compact the knowledge base to a set of more abstract rules. After giving a definition of Policy Propagation Rule, we show (1) a methodology to abstract policy propagation rules based on an ontology, (2) how effective this methodology is when using the Datanode ontology, (3) how this ontology can evolve in order to better represent the behaviour of policy propagation rules.

In this paper, we propose an ontology design pattern for the concept of "explanation". The motivation behind this work comes from our research, which focuses on automatically identifying explanations for data patterns. If we want to produce explanations from data agnostically from the application domain, we first need a formal definition of what an explanation is, i.e. which are its components, their roles or their interactions. We analysed and surveyed works from the disciplines grouped under the name of Cognitive Sciences, with the aim of identifying differences and commonalities in the way their researchers intend the concept of explanation. We then produced not only an ontology design pattern to model it, but also the instantiations of this in each of the analysed disciplines. Besides those contributions, the paper presents how the proposed ontology design pattern can be used to analyse the validity of the explanations produced by our, and other, frameworks.

Most Knowledge Representation (KR) research follows a topdown approach: i) formalisms are designed on the basis of modelling needs and computational considerations, and ii) tools and applications based on these formalisms are realized and tested on application domains. As a result, there has traditionally been little attention in the KR research community to user issues, in particular to the usability of alternative modelling solutions. When statements about the intuitiveness of different solutions are found in the literature, these tend to reflect an author's epistemological standpoint, rather than any concrete user experience. In this paper we take a bottom-up, user-centric perspective and we report on an empirical study where subjects have been asked to represent temporal information and have been provided with alternative design patterns to do so. The study shows that, depending on their experience and level of expertise in KR, users tend to select different patterns for the given modelling problems. In particular, experts appear to choose on the basis of representation power, while naïve users appear to select on the basis of surface features and perceived user-friendliness. Interestingly, while some patterns are indeed perceived to be more intuitive than others, these considerations seem to apply primarily to less experienced users. Indeed, our findings appear to indicate that experts consider issues of 'intuitiveness' as secondary and, in contrast with naïve users, may be happy to apply patterns, which can be regarded as counter-intuitive, if they provide the right tool for the job.

This work investigates the process of selecting, extracting and reorganizing content from Semantic Web information sources, to produce an ontology meeting the specifications of a particular domain and/or task. The process is combined with traditional text-based ontology learning methods to achieve tolerance to knowledge incompleteness. The paper describes the approach and presents experiments in which an ontology was built for a diet evaluation task. Although the example presented concerns the specific case of building a nutritional ontology, the methods employed are domain independent and transferrable to other use cases.

With the rise of linked data, more and more semantically described information is being published online according to the principles and technologies of the Semantic Web (especially, RDF and SPARQL). The use of such standard technologies means that this data should be exploitable, integrable and reusable straight away. However, once a potentially interesting dataset has been discovered, significant efforts are currently required in order to understand its schema, its content, the way to query it and what it can answer. In this paper, we propose a method and a tool to automatically discover questions that can be answered by an RDF dataset. We use formal concept analysis to build a hierarchy of meaningful sets of entities from a dataset. These sets of entities represent answers, which common characteristics represent the clauses of the corresponding questions. This hierarchy can then be used as a querying interface, proposing questions of varying levels of granularity and specificity to the user. A major issue is however that thousands of questions can be included in this hierarchy. Based on an empirical analysis and using metrics inspired both from formal concept analysis and from ontology summarization, we devise an approach for identifying relevant questions to act as a starting point to the navigation in the question hierarchy.

Search in folksonomies is hampered by the fact that the meaning of tags and their relations are not made explicit in the system. This is typically addressed by using knowledge sources (KS) to semantically enrich tagspaces, most notably WordNet and (online) ontologies. However, there is no insight of how the different characteristics of these KS contribute to search improvement in folksonomies. In this work we compare these two KS in the context of folksonomy search. We show that while WordNet leads to richer tag structures than online ontologies do, its fine-grained sense hierarchy renders these structures less effective in search compared to the ones generated from ontologies.

PowerAqua is a Question Answering system, which takes as input a natural language query and is able to return answers drawn from relevant semantic resources found anywhere on the Semantic Web. In this paper we provide two novel contributions: First, we detail a new component of the system, the Triple Similarity Service, which is able to match queries effectively to triples found in different ontologies on the Semantic Web. Second, we provide a first evaluation of the system, which in addition to providing data about PowerAqua's competence, also gives us important insights into the issues related to using the Semantic Web as the target answer set in Question Answering. In particular, we show that, despite the problems related to the noisy and incomplete conceptualizations, which can be found on the Semantic Web, good results can already be obtained.

Because poor quality semantic metadata can destroy the effectiveness of semantic web technology by hampering applications from producing accurate results, it is important to have frameworks that support their evaluation. However, there is no such framework developedto date. In this context, we proposed i) an evaluation reference model, SemRef, which sketches some fundamental principles for evaluating semantic metadata, and ii) an evaluation framework, SemEval, which provides a set of instruments to support the detection of quality problems and the collection of quality metrics for these problems. A preliminary case study of SemEval shows encouraging results.

We propose a knowledge fusion architecture KnoFuss based on the application of problem-solving methods technology, which allows methods for subtasks of the fusion process to be combined and the best methods to be selected, depending on the domain and task at hand.

In this paper, we present an ontology developed to support annotating and reasoning about philosophical knowledge. By this term, we refer to both the factual domain of philosophers, concerning their lives and publications, and the theoretical domain, concerning the ideas they have produced and their relationships. The ontology provides a novel model, which brings together a number of existing formalizations and a series of new ones. In this paper we describe the design of the ontology, in particular focusing on the solutions we have devised to address the modelling problems and natural language ambiguities, which occur when capturing knowledge in a domain as complex as that of philosophy. This work is being carried out in the context of developing a tool, PhiloSURFical, which aims to support ontology-driven exploration of philosophical resources. Specifically, the ontology makes it possible to navigate the philosophical domain, according to a number of narratives, which include theoretical, historical, argumentational, geographic, and others.

In this paper, we propose a generic library of problem-solving methods for scheduling applications. Although some attempts have been made in the past at developing the libraries of scheduling problem-solvers, these only provide limited coverage. Many lack generality, as they subscribe to a particular scheduling domain. Others simply implement a particular problem-solving technique, which may be applicable only to a subset of the space of scheduling problems. In addition, most of these libraries fail to provide the required degree of depth and precision. In our approach, we subscribe to the task-method-domain-application knowledge modeling framework which provides a structured organization for the different components of the library. At the task level, we construct a generic scheduling task ontology to formalize the space of scheduling problems. At the method level, we construct a generic problem-solving model of scheduling that generalizes from the variety of approaches to scheduling problem-solving, which can be found in the literature. The generic nature of this model is demonstrated by constructing seven methods for scheduling as an alternative specialization of the model. Finally, we validated our library on a number of applications to demonstrate its generic nature and effective support for developing scheduling applications

The realization of the Semantic Web is constrained by a knowledge acquisition bottleneck, i.e. the problem of how to add RDF mark-up to the millions of ordinary web pages that already exist. Information Extraction (IE) has been proposed as a solution to the annotation bottleneck. In the task based evaluation reported here, we compared the performance of users without access to annotation, users working with annotations which had been produced from manually constructed knowledge bases, and users working with annotations augmented using IE. We looked at retrieval performance, overlap between retrieved items and the two sets of annotations, and usage of annotation options. Automatically generated annotations were found to add value to the browsing experience in the scenario investigated.

We present a text mining method called CORDER [4] which discovers social networks from an organization's documents. CORDER finds relations between a target named entity and other named entities which occur with it.

We present CORDER (Community Relation Discovery by named Entity Recognition) an un-supervised machine learning algorithm that exploits named entity recognition and co-occurrence data to associate individuals in an organization with their expertise and associates. We discuss the problems associated with evaluating unsupervised learners and report our initial evaluation experiments.

OntoWeaver is our conceptual modelling methodology and a tool that support the specification and implementation of customized web applications. It relies on a number of different types of ontologies to declaratively describe all aspects of a web application. This paper focuses on the OntoWeaver customization framework, which exploits a user model, a customization rule model, and a declarative site model, to enable the design and development of customized web applications at a conceptual level. OntoWeaver makes use of the Jess inference engine to reason upon the site specifications and their underlying site ontologies according to the customization rules and the valuable user profiles to provide customization support in an intelligent way. The ontology-based approach enables the target web applications to be represented in an exchangeable format. Hence, the management and maintenance of web applications can be carried out at a conceptual level without having to worry about the implementation details. Likewise, the declarative nature of the site specifications and the generic customization framework allow the specification of customization requirements to be carried out at the conceptual level.

In this paper we describe a generic library of problem-solving methods (PSMs) for scheduling applications. Although, some attempts have been made in the past at developing libraries of scheduling methods, these only provide limited coverage: in some cases they are specific to a particular scheduling domain; in other cases they simply implement a particular scheduling technique; in other cases they fail to provide the required degree of depth and precision. Our library is based on a structured approach, whereby we first develop a scheduling task ontology, and then construct a task-specific but domain independent model of scheduling problem-solving, which generalises from specific approaches to scheduling problem-solving. Different PSMs are then constructed uniformly by specialising the generic model of scheduling problem-solving. Our library has been evaluated on a number of real-life and benchmark applications to demonstrate its generic and comprehensive nature.

The World Wide Web is opening up access to documents and data for scholars. However it has not yet impacted on one of the primary activities in research: assessing new findings in the light of current knowledge and debating it with colleagues. The ClaiMaker system uses a directed graph model with similarities to hypertext, in which new ideas are published as nodes, which other contributors can build on or challenge in a variety of ways by linking to them. Nodes and links have semantic structure to facilitate the provision of specialist services for interrogating and visualizing the emerging network. By way of example, this paper is grounded in a ClaiMaker model to illustrate how new claims can be described in this structured way.

Formative work by Lave and Wenger has articulated how practices emerge through the interplay of informal processes with symbolic codifications and artifacts. In this paper, we describe how ontologies can serve as symbolic tools within a community of practice supporting communication and knowledge sharing. We show that when a community's perspective on an issue is stable, it opens the possibility for introducing knowledge services, based on an ontology co-constructed by knowledge engineers with stakeholders. Using a case study we describe our approach, ontology driven document enrichment, looking at how ontology construction and population can be supported by web based technologies.

Problem solving methods (PSMs) describe the reasoning components of knowledge-based systems as patterns of behavior that can be reused across applications. While the availability of extensive problem solving method libraries and the emerging consensus on problem solving method specification languages indicate the maturity of the field, a number of important research issues are still open. In particular, very little progress has been achieved on foundational and methodological issues. Hence, despite the number of libraries which have been developed, it is still not clear what organization principles should be adopted to construct truly comprehensive libraries, covering large numbers of applications and encompassing both task-specific and task-independent problem solving methods. In this paper, we address these "fundamental" issues and present a comprehensive and detailed framework for characterizing problem solving methods and their development process. In particular, we suggest that PSM development consists of introducing assumptions and commitments along a three-dimensional space defined in terms of problem-solving strategy, task commitments, and domain (knowledge) assumptions. Individual moves through this space can be formally described by means of adapters. In the paper, we illustrate our approach and argue that our architecture provides answers to three fundamental problems related to research in problem solving methods: 1) what is the epistemological structure and what are the modeling primitives of PSMs? 2) how can we model the PSM development process? and 3) how can we develop and organize truly comprehensive and manageable libraries of problem solving methods?.