Cross-Domain RS: Survey

Cross-Domain RS: Survey

1. Abordagens baseadas em Modelo de Usuário

   • Cross-system user modeling and personalization on the social web (Abel et al, 2011)
   • Mediation of user models for enhanced personalization in recommender systems (Berkovsky et al, 2008)
   • A multi-agent smart user model for cross-domain recommender systems (Gonzáles et al, 2005)
   • Semantic modelling of user interests based on cross-folksonomy analysis (Szomszor et al, 2008)
   • Cross domain recommendation based on multi-type media fusion (Tan et al, 2013)
   • Contextualization, User Modeling and Personalization in the Social Web (Doutorado: Abel, 2011)
   • Exploram modelos de usuário para capturar as preferências sobre itens dos domínios envolvidos.

2. Abordagens para estabelecer relações entre características do domínio

   • Cross-domain recommender systems (Cremonesi et al, 2011)
   • Tags as Bridges between Domains: Improving Recommendation with Tag-Induced Cross-Domain Collaborative Filtering (Shi et al, 2011)
   • CrosSing: A framework to develop knowledge-based recommenders in cross domains(Mestrado: Azak, 2010)
   • How to recommend music to film buffs: enabling the provision of recommendations from multiple domains (Doutorado: Loizou, 2009)
   • A generic semantic-based framework for cross-domain recommendation (Tobias et al, 2011)
   • Location-adapted music recommendation using tags (Kaminskas and Ricci, 2011)
   • Normalmente aplicada a relações de domínio baseada em conteúdo
   • A semantic-based framework for building cross-domain networks: Application to item recommendation (Mestrado: Tobías et al, 2013)
   • Building Ontologies for Cross-domain Recommendation on Facial Skin Problem and Related Cosmetics (Moe & Aung, 2014)
     • Um método para criar ontologias aplicas em cross-domain.
   • Context Aware Cross-domain based Recommendation (Moe & Aung, 2014)
     • O contexto é construído a partir das respostas que o usuário dá ao sistema. Por exemplo: a) há manchas brancas no seu rosto? b) O centro delas é escuro?
   • A Framework for Cross-domain Recommendation in Folksonomies (Guo & Chen, 2013)
   • Cold-Start Management with Cross-Domain Collaborative Filtering and Tags (Enrich et al,2013)
   • Exploraram relações explícitas entre características dos domínios.

3. Abordagens baseadas em Transfer Learning

   • Improving Users’ Acceptance in Recommender System (Doutorado: Wei, 2013)
     • Framework que integra informações de redes sociais com dados de cross-domain.
   • Can movies and books collaborate? Cross-domain collaborative filtering for sparsity reduction (Li et al, 2009)
   • Recentemente tem sido aplicadas à Filtragem Colaborativa.
   • Transfer learning for collaborative filtering via a rating-matrix generative model (Li et al, 2009)
   • Transfer learning in collaborative filtering for sparsity reduction (Pan et al, 2010)
   • Multi-domain collaborative filtering (Zhang et al, 2010)
   • Cross-domain Recommendations based on semantically-enhanced User Web Behavior (Doutorado: Hoxha, 2014)
     • Propõe um modelo formal do comportamento de navegação do usuário
     • Desenvolvimento de um mecanismo para garantir a diversidade de recomendações de vários domínios.
     • Um modelo probabilístico multi-relacional usado para facilitar a transferência de conhecimento entre domínios de sistemas CF
     • O autor observou que não há trabalhos que envolvem as seguintes características de um RS: a) CB + CF, b) Incorporação de representação semântica no conteúdo e; c) Recomendações cross-domain
   • Transfer Learning for Content-Based Recommender Systems using Tree Matching (Biadsy et al, 2013)
   • Active Transfer Learning for Cross-System Recommendation (Zhao et al, 2013)
   • TALMUD - Transfer Learning for Multiple Domains (Moreno et al, 2012)
   • Twin Bridge Transfer Learning for Sparse Collaborative Filtering (Shi et al, 2013)

4. Tutorial on Cross-domain Recommender Systems (Cantador & Cremonesi, 2014)

   • Níveis de domínio
     • Atributo
     • Tipo
     • Item
     • Sistema
   • Tarefas de recomendação
     • Multi-domain
     • Linked-domain
     • Cross-domain
   • Técnicas
     • Linking/aggregating knowledge
       • Merging user preferences
       • Mediating user modeling data
       • Combining recommendations
       • Linking domains
     • Sharing/transferring knowledge
       • Sharing latents features
       • Transferring rating patterns

Bilder und Dateien der Mindmap