ML Anthology

A Theoretical Analysis of Query Selection for Collaborative Filtering

Wee Sun Lee, Philip M. Long
COLT 2001 pp. 517-528
doi:10.1007/3-540-44581-1_34 /colt/2001/lee2001colt-theoretical/

Abstract

We consider the problem of determining which of a set of experts has tastes most similar to a given user by asking the user questions about his likes and dislikes. We describe a simple and fast algorithm for a theoretical model of this problem with a provable approximation guarantee, and prove that solving the problem exactly is NP-Hard.

PDF COLT Semantic Scholar

Cite

Text

Lee and Long. "A Theoretical Analysis of Query Selection for Collaborative Filtering." Annual Conference on Computational Learning Theory, 2001. doi:10.1007/3-540-44581-1_34

Markdown

[Lee and Long. "A Theoretical Analysis of Query Selection for Collaborative Filtering." Annual Conference on Computational Learning Theory, 2001.](https://mlanthology.org/colt/2001/lee2001colt-theoretical/) doi:10.1007/3-540-44581-1_34

BibTeX

@inproceedings{lee2001colt-theoretical,
  title     = {{A Theoretical Analysis of Query Selection for Collaborative Filtering}},
  author    = {Lee, Wee Sun and Long, Philip M.},
  booktitle = {Annual Conference on Computational Learning Theory},
  year      = {2001},
  pages     = {517-528},
  doi       = {10.1007/3-540-44581-1_34},
  url       = {https://mlanthology.org/colt/2001/lee2001colt-theoretical/}
}