I do academic research in the Semantic Web, Artificial Intelligence and Internet of Things fields, working with smart people at the Information Systems Laboratory of the Polytechnic University of Bari. Here's a list of my published work.

RideMATCHain: a Semantic-enhanced Blockchain Marketplace for Ridesharing

8th Italian Conference on ICT for Smart Cities and Communities (I-CiTies 2022)
The integration of blockchain platforms with Semantic Web technologies can increase information interoperability and enable knowledge-driven task automation by intelligent software agents. The paper presents a blockchain platform extending Hyperledger Sawtooth with semantic Smart Contracts allowing annotated resource registration, discovery, explanation, and selection. The platform leverages non-standard inferences for semantic matchmaking, endowed with logic-based justification of results. A prototypical semantic-enhanced ridesharing marketplace has been built to validate the proposal.

Cowl: a lightweight OWL library for the Semantic Web of Everything

1st International Workshop on the Semantic Web of Everything (SWEET 2022)
The Semantic Web of Everything, a blend of the emerging Internet of Everything and well-established Semantic Web paradigms, demands efficient cross-platform knowledge management technologies and tools, able to span the wide variety of devices it aims to support. This paper presents Cowl, a C library for processing Web Ontology Language (OWL) 2 ontologies, designed for strict portability and efficiency constraints. Its architecture is described and main optimization strategies are outlined. Results of a preliminary experimental campaign validate its effectiveness by comparing it with other state-of-the-art OWL toolkits.

Osmotic Cloud-Edge Intelligence for IoT-Based Cyber-Physical Systems

Artificial Intelligence in Cyber-Physical Systems allows machine learning inference on acquired data with ever greater accuracy, thanks to models trained with massive amounts of information generated by IoT devices. Edge Intelligence is increasingly adopted to execute inference on data at the border of local networks, exploiting models trained in the Cloud. However, training tasks on Edge nodes are not yet capable of a flexible, dynamic migration between Edge and Cloud. This paper proposes a Cloud-Edge AI microservice architecture, based on Osmotic Computing principles, which enables training and inference on the Edge, Cloud, or both, exploiting computational resources opportunistically to reach the best prediction accuracy. Microservice encapsulation of each architectural module allows a direct mapping with COTS components. Grounding on the proposed architecture, a prototype has been realized with commodity hardware leveraging open-source software technologies, and it has then been used in a small-scale intelligent manufacturing case study, carrying out experiments that validate the feasibility and key benefits of the approach.

A multiplatform reasoning engine for the Semantic Web of Everything

Journal of Web Semantics
The Internet of Everything and Semantic Web can be joined by giving more intelligence to pervasive systems. To that end, reasoning capabilities should be enabled even for very resource-constrained embedded devices. This paper presents Tiny-ME (the Tiny Matchmaking Engine), a matchmaking and reasoning engine for the Web Ontology Language (OWL), designed and implemented with a compact and portable C core. Main features are high resource efficiency and multiplatform support, spanning containerized microservices, desktops, mobile devices, and embedded boards. The OWLlink interface has been extended to enable non-standard reasoning services for matchmaking in Web, Cloud, and Edge computing. A prototype evaluation is proposed, including a case study on the Pixhawk Unmanned Aerial Vehicle (UAV) autopilot and performance highlights.

A multiplatform energy-aware OWL reasoner benchmarking framework

Journal of Web Semantics
Performance evaluation is increasingly relevant for Web Ontology Language (OWL) reasoners, due to the expanding availability of knowledge corpuses on the Web, the growing variety of applications, and the rise to prominence of mobile and pervasive computing. Motivated mainly by the difficulty of comparing reasoning engines in the Semantic Web of Things (SWoT), this paper introduces evOWLuator, a novel approach and a multiplatform framework devised to be both flexible and expandable. It features integration of traditional and mobile/embedded engines as well as ontology dataset management, reasoning test execution, and report generation. A case study consisting of an experimental setting for time, memory peak and energy footprint evaluation with eight reasoners and four different platforms allows showcasing usage and validating features and usability of the tool.

Mini-ME Swift: the first mobile OWL reasoner for iOS

16th Extended Semantic Web Conference (ESWC 2019)
Mobile reasoners play a pivotal role in the so-called Semantic Web of Things. While several tools exist for the Android platform, iOS has been neglected so far. This is due to architectural differences and unavailability of OWL manipulation libraries, which make porting existing engines harder. This paper presents Mini-ME Swift, the first Description Logics reasoner for iOS. It implements standard (Subsumption, Satisfiability, Classification, Consistency) and non-standard (Abduction, Contraction, Covering, Difference) inference tasks in an OWL 2 language fragment. Peculiarities and optimization are discussed and performance results are presented, comparing Mini-ME Swift with other state-of-the-art OWL reasoners.

Reti veicolari basate sulla rappresentazione della conoscenza

Ital-IA 2019 - Convegno Nazionale CINI sull'Intelligenza Artificiale
Il contributo presenta la ricerca condotta nell’ambito dei sistemi avanzati di assistenza alla guida basati sul paradigma del Semantic Web of Things (SWoT). L’intelligenza artificiale è impiegata per: (i) annotare dati di funzionamento di un veicolo estratti dalle unità di controllo, mediante linguaggi di rappresentazione della conoscenza; (ii) elaborarli combinando ragionamento automatico e machine learning per la rilevazione del contesto e dei fattori di rischio; (iii) disseminare e fondere in maniera intelligente la conoscenza inferita all’interno di una rete veicolare.


Youth Solutions Report 2018 - United Nations "Sustainable Development Solutions Network" (SDSN)
Technical skills and resources of farmers in Tanzania are often insufficient to warrant satisfactory income levels, as well as good sustainability in coffee production. The "HowtUyoga" project proposes an entrepreneurial training course aimed at transferring knowledge and expertise related to the creation and management of a plant that would allow the reuse of organic waste resulting from coffee farming to produce mushroom self-cultivation kits. The goal is to establish virtuous dynamics of environmental, social and economic sustainability (2030 Agenda) while also involving trainers, entrepreneurs, financing institutions and other stakeholders.

OWL API for iOS: early implementation and results

13th OWL: Experiences and Directions Workshop and 5th OWL reasoner evaluation workshop (OWLED - ORE 2016)
Semantic Web and Internet of Things are progressively converging, but the lack of reasoning tools for mobile devices on the iOS platform may hinder the progress of this vision. The paper presents an early redesign of OWL API for iOS. A partial port has been developed, effective enough to support mobile reasoning engines in a moderately expressive fragment of OWL 2. Both architecture and mobile-oriented optimizations are discussed and preliminary performance results are sketched.