Knowledge Graphs provide a machine-readable representation of knowledge conforming to graph-based data models. Link prediction methods predict missing facts in incomplete knowledge graphs, often using scalable embedding based solutions that, however, lack comprehensibility which is crucial in many domains. Filling this gap, explanation methods identify supporting knowledge. For evaluating them, user studies are the obvious choice as users are the main recipients of explanations. However, finding domain experts is often challenging. In contrast, an automated approach is to measure the influence of explanations on the very same link prediction task, thus disregarding the perspective of users. Additionally, current evaluation methods vary across different explanation approaches. We propose LP-DIXIT, the first protocol to evaluate the utility of explanations of link predictions. LP-DIXIT is user-aware, algorithmic and unique for different explanation methods. It builds on a typical setting of user studies, but adopts Large Language Models (LLMs) to mimic users. Specifically, it measures how explanations improve the user (LLM) ability to perform predictions, which is key to trust. We experimentally proved an overall agreement between LP-DIXIT and user evaluations. Moreover, we adopted LP-DIXIT to conduct a comparative study of state-of-the-art explanation methods. The outcomes suggest that less is more: the most effective explanations are those consisting of a single fact.

With the recent surge in popularity of Large Language Models (LLMs), there is the rising risk of users blindly trusting the information in the response. Nevertheless, there are cases where the LLM recommends actions that have potential legal implications and this may put the user in danger. We provide an empirical analysis on multiple existing LLMs showing the urgency of the problem. Hence, we propose a first short-term solution, consisting in an approach for isolating these legal issues through prompt engineering. We prove that this solution is able to stem some risks related to legal implications, nonetheless we also highlight some limitations. Hence, we argue on the need for additional knowledge-intensive resources and specifically Knowledge Graphs for fully solving these limitations. For the purpose, we draw our proposal aiming at designing and developing a solution powered by a legal Knowledge Graph (KG) that, besides capturing and alerting the user on possible legal implications coming from the LLM answers, is also able to provide actual evidence for them by supplying citations of the interested laws. We conclude with a brief discussion on the issues that may be needed to solve for building a comprehensive legal Knowledge Graph

We present a method for boosting relational classifiers of individual resources in the context of the Web of Data . We show how weak classifiers induced by simple concept learners can be enhanced producing strong classification models from training datasets. Even more so the comprehensibility of the model is to some extent preserved as it can be regarded as a sort of concept in disjunctive form. We demonstrate the application of this approach to a weak learner that is easily derived from learners that search a space of hypotheses, requiring an adaptation of the underlying heuristics to take into account weighted training examples. An experimental evaluation on a variety of artificial learning problems and datasets shows that the proposed approach enhances the performance of the basic learners and is competitive, outperforming current concept learning systems.

We focus on the problem of predicting missing assertions in Web ontologies. We start from the assumption that individual resources that are similar in some aspects are more likely to be linked by specific relations: this phenomenon is also referred to as homophily and emerges in a variety of relational domains. In this article, we propose a method for (1) identifying which relations in the ontology are more likely to link similar individuals and (2) efficiently propagating knowledge across chains of similar individuals. By enforcing sparsity in the model parameters, the proposed method is able to select only the most relevant relations for a given prediction task. Our experimental evaluation demonstrates the effectiveness of the proposed method in comparison to state-of-the-art methods from the literature.

We consider the problem of predicting missing class-memberships and property values of individual resources in Web ontologies. We first identify which relations tend to link similar individuals by means of a finite-set Gaussian Process regression model, and then efficiently propagate knowledge about individuals across their relations. Our experimental evaluation demonstrates the effectiveness of the proposed method.

We present a classification method, founded in the instance-based learning and the disjunctive version space approach, for performing approximate retrieval from knowledge bases expressed in Description Logics. The method supplies answers even if the knowledge base of reference is inconsistent or incomplete. Moreover, the method may also induce new knowledge that can be suggested to the knowledge engineer, thus making the ontology population task semi-automatic.

The tasks of resource classification and retrieval from knowledge bases in the Semantic Web are the basis for a lot of important applications. In order to overcome the limitations of purely deductive approaches to deal with these tasks, inductive (instance-based) methods have been introduced as efficient and noise-tolerant alternatives. In this paper we propose an original method based on a non-parametric learning scheme: the Reduced Coulomb Energy (RCE) Network. The method requires a limited training effort but it turns out to be very effective during the classification phase. Casting retrieval as the problem of assessing the classmembership of individuals w.r.t. the query concepts, we propose an extension of a classification algorithm using RCE networks based on an entropic similarity measure for OWL. Experimentally we show that the performance of the resulting inductive classifier is comparable with the one of a standard reasoner and often more efficient than with other inductive approaches. Moreover, we show that new knowledge (not logically derivable) is induced and the likelihood of the answers may be provided.

We present a method based on clustering techniques to detect possible/probable novel concepts or concept drift in a Description Logics knowledge base. The method exploits a semi-distance measure defined for individuals, that is based on a finite number of dimensions corresponding to a committee of discriminating features (concept descriptions). A maximally discriminating group of features is obtained with a randomized optimization method. In the algorithm, the possible clusterings are represented as medoids (w.r.t. the given metric) of variable length. The number of clusters is not required as a parameter, the method is able to find an optimal choice by means of evolutionary operators and a proper fitness function. An experimentation proves the feasibility of our method and its effectiveness in terms of clustering validity indices. With a supervised learning phase, each cluster can be assigned with a refined or newly constructed intensional definition expressed in the adopted language.

We present an evolutionary clustering method which can be applied to multi-relational knowledge bases storing semantic resource annotations expressed in the standard languages for the Semantic Web. The method exploits an effective and language-independent semi-distance measure defined for the space of individual resources, that is based on a finite number of dimensions corresponding to a committee of features represented by a group of concept descriptions (discriminating features). We show how to obtain a maximally discriminating group of features through a feature construction method based on genetic programming. The algorithm represents the possible clusterings as strings of central elements (medoids, w.r.t. the given metric) of variable length. Hence, the number of clusters is not needed as a parameter since the method can optimize it by means of the mutation operators and of a proper fitness function. We also show how to assign each cluster with a newly constructed intensional definition in the employed concept language. An experimentation with some ontologies proves the feasibility of our method and its effectiveness in terms of clustering validity indices.