TransRank: A Novel Algorithm for Transfer of Rank Learning

  title={TransRank: A Novel Algorithm for Transfer of Rank Learning},
  author={Depin Chen and Jun Yan and G. Wang and Y. Xiong and W. Fan and Z. Chen},
  journal={2008 IEEE International Conference on Data Mining Workshops},
  • Depin Chen, Jun Yan, +3 authors Z. Chen
  • Published 2008
  • Computer Science
  • 2008 IEEE International Conference on Data Mining Workshops
  • Recently, learning to rank technique has attracted much attention. However, the lack of labeled training data seriously limits its application in real-world tasks. In this paper, we propose to break this bottleneck by considering the cross-domain ldquotransfer of rank learningrdquo problem. Simultaneously, we propose a novel algorithm called TransRank, which can effectively utilize the labeled data from a source domain to enhance the learning of ranking function in the target domain. The… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Learning to rank with partially-labeled data
    • 99
    • PDF
    Knowledge transfer for cross domain learning to rank
    • 46
    Ranking Model Adaptation for Domain-Specific Search
    • 49
    • PDF
    Learning to rank only using training data from related domain
    • 42
    Personalized ranking model adaptation for web search
    • 65
    • Highly Influenced
    • PDF
    Relevant knowledge helps in choosing right teacher: active query selection for ranking adaptation
    • 22
    • PDF
    Semi-supervised ranking for document retrieval
    • 23
    • PDF
    Pairwise cross-domain factor model for heterogeneous transfer ranking
    • 8
    • PDF
    A Novel Framework for Ranking Model Adaptation
    • 2


    Publications referenced by this paper.
    Learning to rank: from pairwise approach to listwise approach
    • 1,428
    • PDF
    AdaRank: a boosting algorithm for information retrieval
    • 757
    • Highly Influential
    • PDF
    An Efficient Boosting Algorithm for Combining Preferences
    • 2,034
    • PDF
    LETOR: Benchmark Dataset for Research on Learning to Rank for Information Retrieval
    • 489
    • Highly Influential
    • PDF
    Domain Adaptation with Structural Correspondence Learning
    • 1,285
    • Highly Influential
    • PDF
    Learning to Rank with Nonsmooth Cost Functions
    • 229
    McRank: Learning to Rank Using Multiple Classification and Gradient Boosting
    • 378
    • PDF
    Learning to rank using gradient descent
    • 2,098
    • PDF
    Co-clustering based classification for out-of-domain documents
    • 320
    • PDF
    Frustratingly Easy Domain Adaptation
    • 1,433
    • Highly Influential
    • PDF