Research Areas
Research Areas
Embedded Files
Beneath the complexities of modern life lie floods of interconnected information that can be modeled as graphs. The goal of the KDG Lab is to discover knowledge from graph-shaped data. On the one hand, we focus on Knowledge graphs (KGs) made of different types of nodes (e.g., users, movies) and edges (e.g., friend, directed) that offer a natural way to model and capture the complexity and interaction of different entity types. We carry out research in the areas of query languages, reasoning, and representation learning. On the other hand, we focus on social networks and, in particular, on problems related to the privacy of online users that can be compromised by network analysis tools.
Beneath the complexities of modern life lie floods of interconnected information that can be modeled as graphs. The goal of the KDG Lab is to discover knowledge from graph-shaped data. On the one hand, we focus on Knowledge graphs (KGs) made of different types of nodes (e.g., users, movies) and edges (e.g., friend, directed) that offer a natural way to model and capture the complexity and interaction of different entity types. We carry out research in the areas of query languages, reasoning, and representation learning. On the other hand, we focus on social networks and, in particular, on problems related to the privacy of online users that can be compromised by network analysis tools.
Knowledge Graphs
Knowledge Graphs
Topics
Topics
Explanations
Querying
Fact Checking
Publications
V. Fionda, G. Pirrò, M. Consens. Querying Knowledge Graphs with Extended Property Paths. Semantic Web Journal (SWJ), vol. 10, no. 6, pp. 1127-1168, 2019.
G. Pirrò. Building Relatedness Explanations from Knowledge Graphs. Semantic Web Journal (SWJ), vol. 10, no. 6, pp. 963-990, 2019.
F. Darari, W. Nutt, G. Pirrò, S. Razniewski. Completeness Management for RDF Data Sources. ACM Transactions on the Web (TWEB), vol. 13, n. 3, Article No. 18, 2018.
. Fionda, G. Pirrò. Explaining and Querying Knowledge Graphs by Relatedness. 43rd International Conference on Very Large Databases (PVLDB), vol. 10, n. 12, pp. x-y, August 2017.
M. W. Chekol, G. Pirrò, J. Schönfisch, H. Stuckenschmidt. TeCoRe: Temporal Conflict Resolution in Knowledge Graphs. 43rd International Conference on Very Large Databases (PVLDB), vol. 10, n. 12, pp. x-y, August 2017.
V. Fionda, G. Pirrò, C. Gutierrez. Building Knowledge Maps of Web Graphs. Artificial Intelligence (AIJ), pp. 143-167, 2016.
O. Hartig, G. Pirrò. SPARQL with Property Paths on the Web. Semantic Web Journal (SWJ) 8(6), pp. 773-795, 2017.
M. Consens, V. Fionda, S. Khatchadourian and G. Pirrò. S+EPP: Construct and Explore Bisimulation Summaries, plus Optimize Navigational Queries; all on Existing SPARQL Systems. 41st International Conference on Very Large Databases (PVLDB), vol 8, n.12, pp. 2028-2031, September 2015.
V. Fionda, G. Pirrò, C. Gutierrez. NautiLOD: A Formal Language for the Web of Data Graph. ACM Transactions on the Web (TWEB), vol. 9, n. 1, Article No. 5, 2015.
V. Fionda, C. Gutierrez, G. Pirrò. The swget portal: Navigating and Acting on the Web of Linked Data. Journal of Web Semantics (JWS), vol. 26, 2014.
G. Pirrò. Relatedness and TBox-Driven Rule Learning in Large Knowledge Bases. Proceedings of the 34th Conference on Artificial Intelligence (AAAI), New York, Usa, AAAI Press, pp. x-y. February 2020. Acceptance rate: ~20%. GII-GRIN-SCIE Conference Rating A++.
V. Fionda, G. Pirrò. Fact Checking via Evidence Patterns. Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI), Stockholm, Sweden, July 2018. Acceptance rate (track): ~16.5%. GII-GRIN-SCIE Conference Rating A++. Slides available here.
V. Fionda, G. Pirrò. Metastructures in Knowledge Graphs. Proceedings of the 16th International Semantic Web Conference (ISWC), Vienna, Austria, pp. 296-312, LNCS 10587, Springer International Publishing Switzerland, October 2017. Acceptance rate: ~22%. GII-GRIN-SCIE Conference Rating A+. Slides available here.
V. Fionda, G. Pirrò. Explaining Graph Navigational Queries. Proceedings of the 14th European Semantic Web Conference (ESWC), Portoroz, Slovenia, pp. 19-34, LNCS 10249, Springer International Publishing Switzerland, June 2017. Acceptance rate: ~25%. GII-GRIN-SCIE Conference Rating A. Slides available here.
M. W. Chekol, G. Pirrò, J. Schönfisch, H. Stuckenschmidt. Marrying Uncertainty and Time in Knowledge Graphs. Proceedings of the 31st Conference on Artificial Intelligence (AAAI), San Francisco, California, Usa, AAAI Press, pp. 88-94. February 2017. Acceptance rate: ~25%. GII-GRIN-SCIE Conference Rating A++.
M. W. Chekol, G. Pirrò. Containment of Expressive SPARQL Navigational Queries. Proceedings of the 15th International Semantic Web Conference (ISWC), Kobe, Japan, pp. 86-101, LNCS 9981, Springer-Verlag, October 2016. Acceptance rate: ~18%. GII-GRIN-SCIE Conference Rating A+.
G. Pirrò, A. Cuzzocrea. RECAP: Building Relatedness Explanations on the Web. Proceedings of the 25th World Wide Web Conference (WWW); Demo Paper, Montreal, Canada, pp. 235-238, ACM, May 2016. GII-GRIN-SCIE Conference Rating A++.
G. Pirrò. Explaining and Suggesting Relatedness in Knowledge Graphs. Proceedings of the 14th International Semantic Web Conference (ISWC), Bethlehem, Pennsylvania, USA, pp. 622-639, LNCS 9366, Springer-Verlag, October 2015. Acceptance rate: ~22%. Slides available here. GII-GRIN-SCIE Conference Rating A+.
O. Hartig, G. Pirrò. A Context-Based Semantics for SPARQL Property Paths over the Web. Proceedings of the 12th European Semantic Web Conference (ESWC), Portoroz, Slovenia, Springer International Publishing Switzerland, LNCS 9088, pp. 71-87, June 2015. Acceptance rate: ~23%. Best Research Paper Award. GII-GRIN-SCIE Conference Rating A.
V. Fionda, G. Pirrò, M. Consens. Extended Property Paths: Writing More SPARQL Queries in a Succinct Way. Proceedings of the 29th Conference on Artificial Intelligence (AAAI), Austin, Texas, Usa, January 2015. Acceptance rate: ~26%. GII-GRIN-SCIE Conference Rating A++.
Graph Representation Learning
Graph Representation Learning
Triple embedding
GNN
Overlay GNN
Publications
G. Pirrò. LoGNet: Local and Global Triple Embedding Network. Proceedings of the 21st International Semantic Web Conference (ISWC), pp. x-x, LNCS xxx, Springer International Publishing Switzerland, October 2022. Acceptance rate: ~20%. GII-GRIN-SCIE Conference Rating A+.
V. Fionda, G. Pirrò. Learning Triple Embeddings from Knowledge Graphs. Proceedings of the 34th Conference on Artificial Intelligence (AAAI), New York, USA, AAAI Press, pp. x-y. February 2020. Acceptance rate: ~20%. GII-GRIN-SCIE Conference Rating A++. Slides are available here.
Community Deception in Social Networks
Community Deception in Social Networks
One of the main tasks that can be performed over social networks is community detection, that is, identifying a (non-overlapping) partition of nodes of the network and providing some insights about its structure. Although community detection has received much attention, the question concerning what disclosing the community structure of networks can cause to the users remains unsolved. As an example, information about users’ community members can be used by governments to block forms of social self-organization. Another example is the case of Bitcoin trading, where communities are used to identify multiple addresses belonging to the same user. These scenarios underline the need to promote (simple) techniques that can be used by the participants to a community that wants to remain below the radar of network analysis techniques like community detection. We study the problem of community hiding or community deception, which is concerned with devising algorithms that, through principled network updates, can hide the community affiliations in order, for instance, to preserve user privacy in social networks like Facebook or Twitter.
One of the main tasks that can be performed over social networks is community detection, that is, identifying a (non-overlapping) partition of nodes of the network and providing some insights about its structure. Although community detection has received much attention, the question concerning what disclosing the community structure of networks can cause to the users remains unsolved. As an example, information about users’ community members can be used by governments to block forms of social self-organization. Another example is the case of Bitcoin trading, where communities are used to identify multiple addresses belonging to the same user. These scenarios underline the need to promote (simple) techniques that can be used by the participants to a community that wants to remain below the radar of network analysis techniques like community detection. We study the problem of community hiding or community deception, which is concerned with devising algorithms that, through principled network updates, can hide the community affiliations in order, for instance, to preserve user privacy in social networks like Facebook or Twitter.
Publications
V. Fionda, S. A. Madi, G. Pirrò. Community Deception: from Undirected to Directed Networks. Social Network Analysis and Mining (SNAM), vol. 12, no. 74, pp. x-x, 2022.
S. A. Madi, G. Pirrò. Influence-based Community Deception. Proceedings of the 11th International Conference on Complex Networks and their Applications (Complex Networks), November 2022, pp. x-y, 2022.
V. Fionda, G. Pirrò. Community Deception in Networks: Where We Are and Where We Should Go. Proceedings of the 10th International Conference on Complex Networks and their Applications (Complex Networks), Studies in Computational Intelligence, pp. 144–155, 2021
V. Fionda, G. Pirrò. Community Deception in Weighted Networks. Proceedings of the 2021 IEEE/ACM conference on Advances in Social Network Analysis and Mining (ASONAM), pp. x-y, ACM, Accepted.
V. Fionda, G. Pirrò. Community Deception or: How to Stop Fearing Community Detection Algorithms. IEEE Transactions on Knowledge and Data Engineering (TKDE), vol. 30, n. 4, pp. 660-673
Page updated
Report abuse