diff options
author | Paul Mineiro <paul-github@mineiro.com> | 2013-12-22 09:23:44 +0400 |
---|---|---|
committer | Paul Mineiro <paul-github@mineiro.com> | 2013-12-22 09:23:44 +0400 |
commit | 6fca859f04a2b4d327d9c09c98229fa99bbd87e1 (patch) | |
tree | 7b999827920aa4cf912bf99ed94906c1d8d0476d /demo | |
parent | c50981a84dd0ff9d10aa70bd4f9122fdb5ee8f24 (diff) |
movielens demo
Diffstat (limited to 'demo')
-rwxr-xr-x | demo/movielens/README.md | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/demo/movielens/README.md b/demo/movielens/README.md index 89416d3e..9a92d98a 100755 --- a/demo/movielens/README.md +++ b/demo/movielens/README.md @@ -2,7 +2,7 @@ Low rank quadratic demo ------------------------------- This demo shows a low-rank approximation to an interaction design matrix -for the [movielens-1M][http://files.grouplens.org/papers/ml-10m-README.html] +for the [movielens-1M](http://files.grouplens.org/papers/ml-10m-README.html) dataset. ### About Low-Rank Interaction Approximations ### @@ -17,9 +17,9 @@ in non-recommendation settings, e.g., when the interaction between two high cardinality categorical variables is desired but the available data is too sparse to learn a full interaction model. -There is a great piece of software called [libfm][http://www.libfm.org/] +There is a great piece of software called [libfm](http://www.libfm.org/) whose raison d'etre is to fit low-rank approximations to interaction -designs, and the main author [Steffen Rendle][http://www.kaggle.com/users/25112/steffen-rendle] does quite well on Kaggle. Imitation is the best form +designs, and the main author [Steffen Rendle](http://www.kaggle.com/users/25112/steffen-rendle) does quite well on Kaggle. Imitation is the best form of flattery. ### Instructions ### |