Search

In this paper, we present a preliminary approach that uses a set of NLP and Deep Learning methods for extracting entities and relationships from research publications and then integrates them in a Knowledge Graph. More specifically, we (i) tackle the challenge of knowledge extraction by employing several state-of-the-art Natural Language Processing and Text Mining tools, (ii) describe an approach for integrating entities and relationships generated by these tools, and (iii) analyse an automatically generated Knowledge Graph including 10,425 entities and 25,655 relationships in the field of Semantic Web.

This paper describes MantisTable, an open source Semantic Table Interpretation tool, which automatically annotates tables using a Knowledge Graph. MantisTable provides a graphical interface allowing users to analyse the results of the semantic table interpretation process and validate the final annotations. The tool also provides a guided mode for viewing and editing annotations by users. Thanks to MantisTable features, it is possible to create semantic annotations and favour the publication and exchange of tabular data.

Neuro-symbolic integration is a current field of investigation in which symbolic approaches are combined with deep learning ones. In this work we start from simple non-relational knowledge that can be extracted from text by considering the co-occurrence of entities inside textual corpora; we show that we can easily integrate this knowledge with Logic Tensor Networks (LTNs), a neuro-symbolic model. Using LTNs it is possible to integrate axioms and facts with commonsense knowledge represented in a sub-symbolic form in one single model performing well in reasoning tasks. In spite of some current limitations, we show that results are promising.

Nowadays, a website is used to disseminate information about an event (e.g., location, dates, time). In the academic world, it is common to develop a website for an event, such as workshops or conferences. Aligning with the “Web of data”, its dissemination should also happen by publishing the information of the event as a knowledge graph, e.g., via RDF that is available through a SPARQL endpoint or a Triple Patterns Fragment server. However, the RDF generation and website development is not always straightforward and can be time-consuming. In this demo, we present the Semantic Academic-event Dissemination (SAD) Generator for generating RDF and websites for academic events. The generator allows to (i) annotate CSV files that contain academic event data and use the annotations to generate a knowledge graph and (ii) generate a website with the information for the event querying the knowledge graph. We used our generator to generate the RDF and website of a real workshop, the KGB workshop. It can be easily reused by organizers of other academic events by simply providing the event’s information in CSV files.

Vocabularies are used for modeling data in Knowledge Graphs (KGs) like the Linked Open Data Cloud and Wikidata. During their lifetime, vocabularies are subject to changes. New terms are coined, while existing terms are modified or deprecated. We first quantify the amount and frequency of changes in vocabularies. Subsequently, we investigate to which extend and when the changes are adopted in the evolution of KGs. We conduct our experiments on three large-scale KGs: the Billion Triples Challenge datasets, the Dynamic Linked Data Observatory dataset, and Wikidata. Our results show that the change frequency of terms is rather low, but can have high impact due to the large amount of distributed graph data on the web. Furthermore, not all coined terms are used and most of the deprecated terms are still used by data publishers. The adoption time of terms coming from different vocabularies ranges from very fast (few days) to very slow (few years). Surprisingly, we could observe some adoptions before the vocabulary changes were published. Understanding the evolution of vocabulary terms is important to avoid wrong assumptions about the modeling status of data published on the web, which may result in difficulties when querying the data from distributed sources.

In recent years, the increasing propagation of hate speech on social media and the urgent need for effective counter-measures have drawn significant investment from governments, companies, as well as empirical research. Despite a large number of emerging scientific studies to address the problem, existing methods are limited in several ways, such as the lack of comparative evaluations which makes it difficult to assess the contribution of individual works. This paper introduces a new method based on a deep neural network combining convolutional and long short term memory networks, and conducts an extensive evaluation of the method against several baselines and state of the art on the largest collection of publicly available datasets to date. We show that our proposed method outperforms state of the art on 6 out of 7 datasets by between 0.2 and 13.8 points in F1. We also carry out further analysis using automatic feature selection to understand the impact of the conventional manual feature engineering process that distinguishes most methods in this field. Our findings challenge the existing perception of the importance of feature engineering, as we show that: the automatic feature selection algorithm drastically reduces the original feature space by over 90% and selects predominantly generic features from datasets; nevertheless, machine learning algorithms perform better using automatically selected features than the original features.

Lean kernels (LKs) are an effective optimization for deriving the causes of unsatisfiability of a propositional formula. Interestingly, no analogous notion exists for explaining consequences of description logic (DL) ontologies. We introduce LKs for DLs using a general notion of consequence-based methods, and provide an algorithm for computing them which incurs in only a linear time overhead. As an example, we instantiate our framework to the DL \({\mathcal {ALC}}\). We prove formally and empirically that LKs provide a tighter approximation of the set of relevant axioms for a consequence than syntactic locality-based modules.