https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/Head https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://www.nanopub.org/nschema#hasAssertion https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/assertion https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://www.nanopub.org/nschema#hasProvenance https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/provenance https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://www.nanopub.org/nschema#hasPublicationInfo https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/pubinfo https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://www.nanopub.org/nschema#Nanopublication https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/assertion https://w3id.org/spaces/iswc/2026/personal-kgs http://purl.org/dc/terms/description Personalized recommendation systems increasingly influence how users discover information, products, services, and opportunities. However, contemporary recommendation approaches often rely on centralized user profiling and opaque machine learning pipelines that limit transparency, user agency, and privacy. Personal Knowledge Graphs (PKGs) provide a Semantic Web-native approach to representing user preferences, behaviors, context, and provenance, enabling recommendation systems that are explainable, interoperable, and user-centric. Recent advances in KG retrieval, Retrieval-Augmented Generation (RAG), federated learning (FL), and large language models (LLMs) have further expanded the possibilities of PKG-driven personalization. This tutorial introduces participants to the emerging area of PKG-powered recommendation systems. Participants will learn how PKGs can serve as semantic foundations for personalization, how decentralized learning architectures preserve user data sovereignty, how KG retrieval can support explainable recommendations, and how LLM-based recommendation systems can be grounded using personal and external KGs. The tutorial combines conceptual presentations with extensive hands-on activities using open-source notebooks, datasets, and software. Participants will construct PKGs, build recommendation pipelines, perform KG retrieval, and explore approaches for mitigating over-personalization and filter bubbles. https://w3id.org/spaces/iswc/2026/personal-kgs http://schema.org/endDate 2026-10-26T18:00:00+01:00 https://w3id.org/spaces/iswc/2026/personal-kgs http://schema.org/location Sezione 8 https://w3id.org/spaces/iswc/2026/personal-kgs http://schema.org/startDate 2026-10-26T14:10:00+01:00 https://w3id.org/spaces/iswc/2026/personal-kgs http://schema.org/url https://brains-group.github.io/PKG-Recs https://w3id.org/spaces/iswc/2026/personal-kgs http://www.w3.org/1999/02/22-rdf-syntax-ns#type https://w3id.org/kpxl/gen/terms/Event https://w3id.org/spaces/iswc/2026/personal-kgs http://www.w3.org/1999/02/22-rdf-syntax-ns#type https://w3id.org/kpxl/gen/terms/Space https://w3id.org/spaces/iswc/2026/personal-kgs http://www.w3.org/1999/02/22-rdf-syntax-ns#type https://w3id.org/scholarlydata/ontology/conference-ontology.owl#Tutorial https://w3id.org/spaces/iswc/2026/personal-kgs http://www.w3.org/2000/01/rdf-schema#label Personal Knowledge Graphs for LLM-Powered Decentralized Recommendations https://w3id.org/spaces/iswc/2026/personal-kgs https://w3id.org/kpxl/gen/terms/hasAdmin https://orcid.org/0000-0002-1267-0234 https://w3id.org/spaces/iswc/2026/personal-kgs https://w3id.org/kpxl/gen/terms/hasAdmin https://w3id.org/np/RA0z3VTaQLNRhIoFyW7C8fJ04tho6xUivHyFqeY-RUv30/event-bot https://w3id.org/spaces/iswc/2026/personal-kgs https://w3id.org/kpxl/gen/terms/hasRootDefinition https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 https://w3id.org/spaces/iswc/2026/personal-kgs https://w3id.org/kpxl/gen/terms/isSubSpaceOf https://w3id.org/spaces/iswc/2026 https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/provenance https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/assertion http://www.w3.org/ns/prov#wasAttributedTo https://w3id.org/np/RA0z3VTaQLNRhIoFyW7C8fJ04tho6xUivHyFqeY-RUv30/event-bot https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/assertion http://www.w3.org/ns/prov#wasDerivedFrom https://iswc2026.semanticweb.org/#/program/schedule https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/assertion http://www.w3.org/ns/prov#wasDerivedFrom https://iswc2026.semanticweb.org/#/program/tutorials https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/pubinfo https://orcid.org/0000-0002-1267-0234 http://xmlns.com/foaf/0.1/name Tobias Kuhn https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://purl.org/dc/terms/created 2026-10-09T07:31:01Z https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://purl.org/dc/terms/creator https://w3id.org/np/RA0z3VTaQLNRhIoFyW7C8fJ04tho6xUivHyFqeY-RUv30/event-bot https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://purl.org/dc/terms/license https://creativecommons.org/licenses/by/4.0/ https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://purl.org/nanopub/x/introduces https://w3id.org/spaces/iswc/2026/personal-kgs https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 http://www.w3.org/2000/01/rdf-schema#label Personal Knowledge Graphs for LLM-Powered Decentralized Recommendations https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 https://w3id.org/np/o/ntemplate/wasCreatedFromProvenanceTemplate https://w3id.org/np/RAefFjYMSnydcZtTqdT8TG8rJeJa76ryG7h10VHQiWKgY https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 https://w3id.org/np/o/ntemplate/wasCreatedFromPubinfoTemplate https://w3id.org/np/RA-N4_kFxcruvBTlo1yJv29E4SLOXsx5KO_th69PprMrA https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 https://w3id.org/np/o/ntemplate/wasCreatedFromPubinfoTemplate https://w3id.org/np/RACJ58Gvyn91LqCKIO9zu1eijDQIeEff28iyDrJgjSJF8 https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 https://w3id.org/np/o/ntemplate/wasCreatedFromPubinfoTemplate https://w3id.org/np/RAYrlN4s93vVe9LGI-gmPLTb-QZGHKd0mxx8VxJ3XVhuw https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 https://w3id.org/np/o/ntemplate/wasCreatedFromPubinfoTemplate https://w3id.org/np/RAoTD7udB2KtUuOuAe74tJi1t3VzK0DyWS7rYVAq1GRvw https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 https://w3id.org/np/o/ntemplate/wasCreatedFromTemplate https://w3id.org/np/RA0hqbtWzJGLLaFG9STgZo6M3-XKssZSSl5RF8T4UMWRg/workshop-template https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/sig http://purl.org/nanopub/x/hasAlgorithm RSA https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/sig http://purl.org/nanopub/x/hasPublicKey MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAp6vyESGMFAOA0EUnUMHGWho/p443sMTrKm+Y1lqZt76Jc9dJggk4NT0dtRt5fZFOEpgDxl6/jyUSRey6d3/LL11iENT1HNgMDRkfRQvojuN8FyYogUf1wBcWk+HeSLCVBAkOnap3ejUPhgIBgUI0w+ElBMagFiORtyDYg9RFElhybOLLQA0Pkj0txN5MMhVnVhp/dzn/4671ogwzVZwejyFRUt0Beh86QS9YD/+ZfOGgYueHOMfUXMulhp13JU3YY0UJ8lLAhnkZEDl0JTeRW3XdsqY42mt8vl4sR2ocR+fUKLSFcdN5qOFt76Pj4cKp/zzjDOUfsVWRLGmpev8qAwIDAQAB https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/sig http://purl.org/nanopub/x/hasSignature AVzH67XH1cDBPPw1Kcjde8iCQoncYtP9ccO/9DDjZq46ckNH724dR7aI4z6rLpPK0I8zdLdSmr1XV60E7G2oqyTAYzz9OLqfQfJKo1LZtvRWBl2bejYmuQNl5HJQU67FX9oZ2C8RsGHhWoKaxkBYl15u2OsVMlj/IsgqAWXD46z9ef1jeyPuEQ8NIsAqa0ojPSN6YNVTSgN8vtULxk+Lhgtkgj7PIHNPCfMvMM7eGEPNGGz32MCRslM/cnZoxP8plDQPauDWCgRP9a6fQ9ibsobljaL9oFF9YP+wQ/NSWperFSHw3FuhJkVkXkyC4tc/W9cIaA/Y5uVBvPmy9i8hpQ== https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/sig http://purl.org/nanopub/x/hasSignatureTarget https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8 https://w3id.org/np/RA67YGEa7SMgk_UOmHFhdMjGp5eHcWMsHp-Uq_rGvm5o8/sig http://purl.org/nanopub/x/signedBy https://w3id.org/np/RA0z3VTaQLNRhIoFyW7C8fJ04tho6xUivHyFqeY-RUv30/event-bot