This course offers an in-depth exploration of knowledge graphs (KGs)—a rapidly growing area of research and application in both industry and academia. Knowledge graphs provide a graph-based abstraction to organize, integrate, and derive value from diverse and complex data sources, forming the backbone of cutting-edge data management and AI systems.
Participants will gain foundational knowledge of KGs, starting with their core principles and extending to advanced techniques in key topics such as:
-
Languages and frameworks for querying knowledge graphs, such as SPARQL.
-
Techniques for data integration, enrichment, and profiling to enhance the utility of KGs.
-
Methods for assessing and refining knowledge graph quality to ensure reliability and accuracy.
-
Emerging intersections of KGs with cutting-edge AI technologies, including retrieval-augmented generation (RAGs) and large language models (LLMs).
The course emphasizes practical applications through real-world case studies, illustrating how knowledge graphs are deployed in domains such as healthcare, finance, and robotics. Participants will engage in hands-on discussions, exploring the current limitations of KGs and their future directions, including scalability, explainability, and integration with evolving AI paradigms.
To ensure accessibility, the course employs illustrative examples, intuitive visualizations, and clear explanations throughout. A portion of the course is reserved for discussions, enabling participants to align course concepts with their research interests and develop domain-specific strategies for KG adoption.
Final Exam:
Students will complete a project or in-depth study focused on the application of a knowledge graph in their specific research domain.
Course Syllabus
Titolo
Introduzione ai Knowledge Graph
Docente(i)
Dott. Marco Cremaschi
Prof.ssa Anisa Rula
Dott.ssa Blerina Spahiu
Lingua
Inglese
Breve descrizione
Questo corso offre un'esplorazione approfondita dei knowledge graph (KG) - un'area di ricerca e applicazione in rapida crescita sia nell'industria che nel mondo accademico. I knowledge graph forniscono un'astrazione basata su grafi per organizzare, integrare e trarre valore da fonti di dati diversificate e complesse, costituendo la spina dorsale dei più avanzati sistemi di gestione dati e intelligenza artificiale.
I partecipanti acquisiranno conoscenze fondamentali sui KGs, a partire dai principi di base fino alle tecniche avanzate in argomenti chiave come:
- Linguaggi e framework per interrogare i knowledge graph, come SPARQL.
- Tecniche per l'integrazione, l'arricchimento e la profilazione dei dati per migliorare l'utilità dei KGs.
- Metodi per valutare e perfezionare la qualità dei knowledge graph al fine di garantirne affidabilità e accuratezza.
- Intersezioni emergenti dei KGs con tecnologie AI all'avanguardia, inclusi Retrieval-Augmented Generation (RAGs) e Large Language Models (LLMs).
Il corso pone l'accento sulle applicazioni pratiche attraverso casi di studio reali, illustrando come i knowledge graph siano impiegati in settori come la sanità, la finanza e la robotica. I partecipanti saranno coinvolti in discussioni interattive per esplorare i limiti attuali dei KGs e le loro direzioni future, comprese scalabilità, spiegabilità e integrazione con paradigmi AI in evoluzione.
Per garantire l'accessibilità, il corso utilizza esempi illustrativi, visualizzazioni intuitive e spiegazioni chiare. Una parte del corso è riservata alle discussioni, permettendo ai partecipanti di allineare i concetti del corso con i propri interessi di ricerca e sviluppare strategie specifiche per l'adozione dei KGs nei propri ambiti.
Esame Finale:
Gli studenti realizzeranno un progetto o uno studio approfondito sull'applicazione di un knowledge graph nel proprio specifico ambito di ricerca.
Programma:
- Introduzione ai Knowledge Graphs (2h)
- Modellazione e ragionamento per i KGs: Vocabolari, Ontologie, RDFS, OWL (2h)
- Linguaggi di interrogazione per i KGs (2h)
- Integrazione e arricchimento dei dati (2h)
- Valutazione e miglioramento della qualità (2h)
- Profilazione dei Knowledge Graph (2h)
- Knowledge Graph, RAGs e LLMs (4h)
- Knowledge Graph nella pratica (4h)
CFU / Ore
20 ore
Periodo di erogazione
Settembre - Ottobre 2025
Title
Introduction to Knowledge Graphs
Teacher(s)
Dott. Marco Cremaschi
Prof. Anisa Rula
Dott. Blerina Spahiu
Language
English
Short description
This course offers an in-depth exploration of knowledge graphs (KGs) - a rapidly growing area of research and application in both industry and academia. Knowledge graphs provide a graph-based abstraction to organize, integrate, and derive value from diverse and complex data sources, forming the backbone of cutting-edge data management and AI systems.
Participants will gain foundational knowledge of KGs, starting with their core principles and extending to advanced techniques in key topics such as:
- Languages and frameworks for querying knowledge graphs, such as SPARQL.
- Techniques for data integration, enrichment, and profiling to enhance the utility of KGs.
- Methods for assessing and refining knowledge graph quality to ensure reliability and accuracy.
Emerging intersections of KGs with cutting-edge AI technologies, including Retrieval-Augmented Generation (RAGs) and Large Language Models (LLMs).
The course emphasizes practical applications through real-world case studies, illustrating how knowledge graphs are deployed in domains such as healthcare, finance, and robotics. Participants will engage in hands-on discussions, exploring the current limitations of KGs and their future directions, including scalability, explainability, and integration with evolving AI paradigms.
To ensure accessibility, the course employs illustrative examples, intuitive visualizations, and clear explanations throughout. A portion of the course is reserved for discussions, enabling participants to align course concepts with their research interests and develop domain-specific strategies for KG adoption.
Final Exam:
Students will complete a project or in-depth study focused on the application of a knowledge graph in their specific research domain.
Syllabus
- Introduction to Knowledge Graphs; (2h)
- Modeling & Reasoning for KGs: Vocabularies, Ontologies, RDFS, OWL (2h)
- Query languages for KGs (2h)
- Data integration and Enrichment (2h)
- Quality Assessment and Refinement (2h)
- Knowledge Graph Profiling (2h)
- Knowledge Graph, RAGs and LLMs (4h)
- Knowledge Graph in practice (4h)
CFU / Hours
20 hours
Teaching period
September - October 2025