Here is a list of some of the main research projects I was involved in the last few years. Most of the projects have received funding either from the EU or from industry.
Marie Skłodowska-Curie personal research fellowship (Sep 2019 - Sep 2021)
KGs are knowledge-bases of facts about entities and concepts (e.g., places, persons, artifacts) which are represented using the flexible structure of a graph. Facts are often extracted from encyclopedic knowledge, such as Wikipedia, or existing structured repositories (e.g. Wikidata), or even from unstructured sources such as social media posts (e.g. Facebook Graph). For example, a KG containing information about organisations is likely to include facts about companies, their founders, key persons, headquarters’ locations, number of employees, etc. However, facts related to entities or concepts that are dynamically changing over time are usually missing or outdated. The evolving dynamics of real world events are usually not reflected into knowledge bases. Hence, current repositories tend to represent only static snapshots of real world entities, ignoring their changes over time. This project aims at exploring solutions for managing temporal and evolving aspects of KGs and leveraging such features for deeper data analysis.
Industry project, in collaboration with Wolters Kluwer, Germany (2018-2019)
Role: Scientific Lead and Scrum Master
The information contained in legal information systems is often accessed through simple keyword-based search interfaces and presented as a simple list of hits. In order to improve search accuracy one may avail of knowledge graphs (KGs), where the semantics of the data can be made explicit. During this project we developed a KG-based search engine designed for the entire corpus of German court cases mantained by Wolters Kluwer, Germany.
Industry project, in collaboration with Huawei Ireland (2018)
Role: Lead Research Scientist
The amount of video content available on the Web is constantly growing. This has made it harder for viewers to discover the right visual content for them. Recommender systems are being offered by VoD services in order to automatically suggest potentially interesting videos to users. However, recommendations are typically based on: (i) limited video metadata fields, such as genre, title and actors; (ii) the content that other users liked; (iii) private or isolated data repositories. Our proposed approach leverages the richness of semantic technologies to enrich movie metadata, and create meaningful descriptions of movie scenes using video and audio processing techniques. This approach allows for the creation of a structured and interoperable Knowledge Graph (KG) describing movies and their content. This enables deeper data analysis of movie content at Web scale including, for example, algorithms for automatic video summarisation.
Role: Work Package & Task Leader
Abstract: Locations that exhibit a certain interest or serve a certain purpose are commonly referred to as Points of Interest (POIs). The concept of a POI is quite broad, encompassing anything from a shop, restaurant or museum to an ATM or bus stop. POI data are the cornerstone of any application, service, and product even remotely related to our physical surroundings. The project delivered the first comprehensive cloud-based platform for the quality-assured world-scale integration of Big POI data assets. SLIPO reduces the effort, time, and complexity of POI data integration, providing POIs of increased size, coverage, richness and timeliness at a fraction of the cost. The SLIPO system enables non-expert of linked data technologies to import, link, fuse, and enrich heterogeneous proprietary and open POI data, regardless of their original format, schema, or identifiers. SLIPO integrates and extends leading open source Linked Data to specifically address the requirements of world-scale POI integration.
Role: Work Package & Task Leader
The main objective of BigDataOcean is to enable maritime big data scenarios for EU-based companies, organisations and scientists, through a multi-segment platform that will combine data of different velocity, variety and volume under an inter-linked, trusted, multilingual engine to produce a big-data repository of value and veracity back to the participants and local communities. In particular I was responsible for WP3 “Cross-Sector Semantics, Analytics and Business Intelligence Algorithms” leading the work on Linked Data vocabularies and the BDO Metadata Repository architecture. In addition to issues of Harmonisation, Knowledge Extraction, Business Intelligence and Usage Analytics Services.
Role: Project Coordinator
Openness and transparency can act as a disincentive to corruption. Government agencies, data wranglers, journalists and even citizens can access a comprehensive online platform to analyse and participate in public budgets. The OpenBudgets.eu project, known as OBEU, has developed an open-source software framework and a software-as-a-service (SAAS) to support financial transparency and enhance accountability within the public sector at all levels up to European level. OBEU has created a Linked Data platform with a complete set of 13 tools, 3 use-cases and many datasets represented as interlinked knowledge graphs.