<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-1653">
  <Title>Relevance Feedback Models for Recommendation</Title>
  <Section position="7" start_page="454" end_page="455" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> Recommender systems help users select particular items from a large number of choices by providing recommendations. Much work has been done to combine content-based filtering (CBF) and collaborative filtering (CF) to provide better recommendations. The contributions reported in this paper are twofold: (1) we extended relevance feed-back approaches to incorporate CF and (2) we introduced the approximated Polya model as a gen- null eralization of the multinomial model and showed that it is better suited to CF and CBF. The performance of the Polya model is comparable to that of a state-of-the-art item-based CF method.</Paragraph>
    <Paragraph position="1"> Our work shows that language modeling approaches in information retrieval can be extended to CF. This implies that a large amount of work in the field of IR could be imported into CF. This would be interesting to investigate in future work.</Paragraph>
  </Section>
class="xml-element"></Paper>