{"id":{"repo_id":"freiburg-diss","oai_identifier":"oai:freidok.uni-freiburg.de:1804"},"canonical_url":"https://search.dev.ndltd.org/etd/freiburg-diss/oai:freidok.uni-freiburg.de:1804","repository":{"repo_id":"freiburg-diss","name":"University of Freiburg","base_url":"https://freidok.uni-freiburg.de/oai/oai2.php"},"display":{"title":"Towards decentralized recommender systems","abstract":"Automated recommender systems make product suggestions that are tailored
to the human user's individual needs and represent powerful means to combat
information glut. However, their practical applicability has been largely confined to scenarios where all information relevant for recommendation making
is kept in one single, authoritative node.
Recently, novel distributed infrastructures are emerging, e.g., peer-to-peer and
ad-hoc networks, the Semantic Web, the Grid, etc., and supersede classical
client/server approaches in many respects. These infrastructures could likewise
benefit from recommender system services, leading to a paradigm shift
towards decentralized recommender systems.
In this thesis, we investigate the challenges that decentralized recommender
systems bring up and propose diverse techniques in order to cope with those
particular issues. The spectrum of methods proposed ranges from the employment
of product classification taxonomies as powerful background knowledge,
alleviating the sparsity problem, to trust propagation mechanisms designed
to address the scalability issue. Empirical investigations on the correlation of
interpersonal trust and interest similarity provide the component glue that
melds these results together and renders the eventual creation of a decentralized recommender framework feasible.
While these building bricks, namely taxonomy-driven filtering, topic diversification, and the Appleseed trust metric, are vital for the conception of our trust-based decentralized recommender, they are also valuable contributions
in their own right, addressing issues not only confined to the universe of decentralized recommender systems.","abstract_html":"Automated recommender systems make product suggestions that are tailored <br>to the human user's individual needs and represent powerful means to combat <br>information glut. However, their practical applicability has been largely confined to scenarios where all information relevant for recommendation making <br>is kept in one single, authoritative node. <br> <br>Recently, novel distributed infrastructures are emerging, e.g., peer-to-peer and <br>ad-hoc networks, the Semantic Web, the Grid, etc., and supersede classical <br>client/server approaches in many respects. These infrastructures could likewise <br>benefit from recommender system services, leading to a paradigm shift <br>towards decentralized recommender systems. <br> <br>In this thesis, we investigate the challenges that decentralized recommender <br>systems bring up and propose diverse techniques in order to cope with those <br>particular issues. The spectrum of methods proposed ranges from the employment <br>of product classification taxonomies as powerful background knowledge, <br>alleviating the sparsity problem, to trust propagation mechanisms designed <br>to address the scalability issue. Empirical investigations on the correlation of <br>interpersonal trust and interest similarity provide the component glue that <br>melds these results together and renders the eventual creation of a decentralized recommender framework feasible. <br> <br>While these building bricks, namely taxonomy-driven filtering, topic diversification, and the Appleseed trust metric, are vital for the conception of our trust-based decentralized recommender, they are also valuable contributions <br>in their own right, addressing issues not only confined to the universe of decentralized recommender systems.","abstract_has_math":false,"creators":["Ziegler, Cai-Nicolas"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Lausen, Georg"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-07-24T02:22:28Z","subjects":["Recommender Systems","Information Filtering","Trust Models","Statistics"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://freidok.uni-freiburg.de/data/1804","outbound_label":"Repository record","outbound_source":"source_url"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Lausen, Georg"]},{"key":"dc:creator","label":"Author","values":["Ziegler, Cai-Nicolas"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:type","label":"Dc Type","values":["DoctoralThesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Recommender Systems","Information Filtering","Trust Models","Statistics"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Automated recommender systems make product suggestions that are tailored
to the human user's individual needs and represent powerful means to combat
information glut. However, their practical applicability has been largely confined to scenarios where all information relevant for recommendation making
is kept in one single, authoritative node.
Recently, novel distributed infrastructures are emerging, e.g., peer-to-peer and
ad-hoc networks, the Semantic Web, the Grid, etc., and supersede classical
client/server approaches in many respects. These infrastructures could likewise
benefit from recommender system services, leading to a paradigm shift
towards decentralized recommender systems.
In this thesis, we investigate the challenges that decentralized recommender
systems bring up and propose diverse techniques in order to cope with those
particular issues. The spectrum of methods proposed ranges from the employment
of product classification taxonomies as powerful background knowledge,
alleviating the sparsity problem, to trust propagation mechanisms designed
to address the scalability issue. Empirical investigations on the correlation of
interpersonal trust and interest similarity provide the component glue that
melds these results together and renders the eventual creation of a decentralized recommender framework feasible.
While these building bricks, namely taxonomy-driven filtering, topic diversification, and the Appleseed trust metric, are vital for the conception of our trust-based decentralized recommender, they are also valuable contributions
in their own right, addressing issues not only confined to the universe of decentralized recommender systems.","Automatisierte Recommender-Systeme berechnen Produktvorschläge, welche
genau auf die Interessen und Bedürfnisse ihrer Benutzer zugeschnitten sind
und stellen somit exzellente Mittel dar, um der stetig wachsenden Informationsflut Herr zu werden. Allerdings sieht sich deren praktische Einsetzbarkeit bis dato weithin auf Szenarien beschränkt, bei denen man alle für die zur Berechnung von Empfehlungen relevante Information als in einem einzigen
Knoten gekapselt annehmen konnte.
Seit einigen Jahren nehmen verteilte Infrastrukturen, wie zum Beispiel Peer-to-
Peer und Ad-Hoc Netzwerke, das Semantic Web, der Grid etc., immer
deutlichere Konturen an und ersetzen klassische Client/Server-Modelle bereits
in vielerlei Hinsicht. Diese Infrastrukturen könnten gleichwohl von den
von Recommender-Systemen bereitgestellten Diensten profitieren und somit
einen Paradigmenwechsel hin zu dezentralisierten Recommender-Systemen
einläuten.
Im Rahmen dieser Dissertation untersuchen wir zunächst die neuen Herausforderungen, denen es sich im Hinblick auf die Konzeption dezentraler
Recommender-Systeme zu stellen gilt, und schlagen diverse neue Ansätze
vor, mit deren Hilfe man speziell jene Probleme zu bewältigen vermag. Das
Spektrum der vorgestellten Methoden reicht dabei von der Verwendung von
mächtigen Taxonomien zur Klassifikation von Produkten zwecks künstlicher
Verdichtung der Daten, bis hin zu Vertrauensmetriken, die entworfen wurden,
um Fragen der Skalierbarkeit derartiger Systeme zu lösen. Empirische Untersuchungen bezüglich der Korrelation interpersonellen Vertrauens und Interessengleichheit stellen den Mörtel dar, welcher jene einzelnen Bausteine zusammenfügt und die schlussendliche Realisierung eines exemplarischen Frameworks für dezentrale Recommender-Systeme ermöglicht.
Während die angesprochenen Bausteine, im namentlichen Taxonomie-basiertes
Filtern, Topic Diversification und die Appleseed Vertrauensmetrik, notwendige
Komponenten für die Konzeption eines auf sozialem Vertrauen basierten,
dezentralen Recommender-Systems darstellen, so sind diese gleichermaßen
wichtige wissenschaftliche Beiträge per se und auch außerhalb der Fragestellung
\"Dezentrale Recommender-Systeme\" von praktischer Relevanz."]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Towards decentralized recommender systems","Dezentrale Empfehlungssysteme"]}]}],"canonical_facts":{"dc:contributor":["Lausen, Georg"],"dc:creator":["Ziegler, Cai-Nicolas"],"dc:description.abstract":["Automated recommender systems make product suggestions that are tailored
to the human user's individual needs and represent powerful means to combat
information glut. However, their practical applicability has been largely confined to scenarios where all information relevant for recommendation making
is kept in one single, authoritative node.
Recently, novel distributed infrastructures are emerging, e.g., peer-to-peer and
ad-hoc networks, the Semantic Web, the Grid, etc., and supersede classical
client/server approaches in many respects. These infrastructures could likewise
benefit from recommender system services, leading to a paradigm shift
towards decentralized recommender systems.
In this thesis, we investigate the challenges that decentralized recommender
systems bring up and propose diverse techniques in order to cope with those
particular issues. The spectrum of methods proposed ranges from the employment
of product classification taxonomies as powerful background knowledge,
alleviating the sparsity problem, to trust propagation mechanisms designed
to address the scalability issue. Empirical investigations on the correlation of
interpersonal trust and interest similarity provide the component glue that
melds these results together and renders the eventual creation of a decentralized recommender framework feasible.
While these building bricks, namely taxonomy-driven filtering, topic diversification, and the Appleseed trust metric, are vital for the conception of our trust-based decentralized recommender, they are also valuable contributions
in their own right, addressing issues not only confined to the universe of decentralized recommender systems.","Automatisierte Recommender-Systeme berechnen Produktvorschläge, welche
genau auf die Interessen und Bedürfnisse ihrer Benutzer zugeschnitten sind
und stellen somit exzellente Mittel dar, um der stetig wachsenden Informationsflut Herr zu werden. Allerdings sieht sich deren praktische Einsetzbarkeit bis dato weithin auf Szenarien beschränkt, bei denen man alle für die zur Berechnung von Empfehlungen relevante Information als in einem einzigen
Knoten gekapselt annehmen konnte.
Seit einigen Jahren nehmen verteilte Infrastrukturen, wie zum Beispiel Peer-to-
Peer und Ad-Hoc Netzwerke, das Semantic Web, der Grid etc., immer
deutlichere Konturen an und ersetzen klassische Client/Server-Modelle bereits
in vielerlei Hinsicht. Diese Infrastrukturen könnten gleichwohl von den
von Recommender-Systemen bereitgestellten Diensten profitieren und somit
einen Paradigmenwechsel hin zu dezentralisierten Recommender-Systemen
einläuten.
Im Rahmen dieser Dissertation untersuchen wir zunächst die neuen Herausforderungen, denen es sich im Hinblick auf die Konzeption dezentraler
Recommender-Systeme zu stellen gilt, und schlagen diverse neue Ansätze
vor, mit deren Hilfe man speziell jene Probleme zu bewältigen vermag. Das
Spektrum der vorgestellten Methoden reicht dabei von der Verwendung von
mächtigen Taxonomien zur Klassifikation von Produkten zwecks künstlicher
Verdichtung der Daten, bis hin zu Vertrauensmetriken, die entworfen wurden,
um Fragen der Skalierbarkeit derartiger Systeme zu lösen. Empirische Untersuchungen bezüglich der Korrelation interpersonellen Vertrauens und Interessengleichheit stellen den Mörtel dar, welcher jene einzelnen Bausteine zusammenfügt und die schlussendliche Realisierung eines exemplarischen Frameworks für dezentrale Recommender-Systeme ermöglicht.
Während die angesprochenen Bausteine, im namentlichen Taxonomie-basiertes
Filtern, Topic Diversification und die Appleseed Vertrauensmetrik, notwendige
Komponenten für die Konzeption eines auf sozialem Vertrauen basierten,
dezentralen Recommender-Systems darstellen, so sind diese gleichermaßen
wichtige wissenschaftliche Beiträge per se und auch außerhalb der Fragestellung
\"Dezentrale Recommender-Systeme\" von praktischer Relevanz."],"dc:format.medium":["application/pdf"],"dc:subject":["Recommender Systems","Information Filtering","Trust Models","Statistics"],"dc:title":["Towards decentralized recommender systems","Dezentrale Empfehlungssysteme"],"dc:type":["DoctoralThesis"]},"updated_at":"2026-07-24T02:22:28Z"}