@incollection{
    Savia04,
    author = {Savia, Eerika and Kaski, Samuel and Tuulos, Ville and Myllym{\"a}ki, Petri},
    publisher = "IEEE",
    title = "On text-based estimation of document relevance",
    url = "http://www.cis.hut.fi/projects/mi/papers/ijcnn04.pdf",
    booktitle = "Proceedings of IJCNN'04, International Joint Conference on Neural Networks",
    address = "Piscataway, NJ",
    abstract = "This work is part of a proactive information retrieval project that aims at estimating relevance from implicit user feedback. The noisy feedback signal needs to be complemented with all available information, and textual content is one of the natural sources. Here we take the first steps by investigating whether this source is at all useful in the challenging setting of estimating the relevance of a new document based on only few samples with known relevance. It turns out that even sophisticated unsupervised methods like multinomial PCA (or Latent Dirichlet Allocation) cannot help much. By contrast, feature extraction supervised by relevant auxiliary data may help.",
    flags = "public",
    year = "2004",
    pages = "3275--3280"
}