Repository: incubator-hivemall Updated Branches: refs/heads/master 072a0586d -> ca9b76a25
Fixed rendering errors in SLIM doc Project: http://git-wip-us.apache.org/repos/asf/incubator-hivemall/repo Commit: http://git-wip-us.apache.org/repos/asf/incubator-hivemall/commit/ca9b76a2 Tree: http://git-wip-us.apache.org/repos/asf/incubator-hivemall/tree/ca9b76a2 Diff: http://git-wip-us.apache.org/repos/asf/incubator-hivemall/diff/ca9b76a2 Branch: refs/heads/master Commit: ca9b76a259c2d039220ce3b14c2298a8cadb3f7f Parents: 072a058 Author: Makoto Yui <m...@apache.org> Authored: Thu Sep 28 12:54:04 2017 +0900 Committer: Makoto Yui <m...@apache.org> Committed: Thu Sep 28 12:54:04 2017 +0900 ---------------------------------------------------------------------- docs/gitbook/recommend/movielens_slim.md | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/ca9b76a2/docs/gitbook/recommend/movielens_slim.md ---------------------------------------------------------------------- diff --git a/docs/gitbook/recommend/movielens_slim.md b/docs/gitbook/recommend/movielens_slim.md index 498e1e6..760f3cd 100644 --- a/docs/gitbook/recommend/movielens_slim.md +++ b/docs/gitbook/recommend/movielens_slim.md @@ -145,7 +145,7 @@ where rnd < 0.5 > we will show the example of queries and its results based on $$K$$-hold > cross validation case. > But, this article's queries are valid for leave-one-out cross validation. -## Precompute movie-movie similarity +## Pre-compute item-item similarity SLIM needs top-$$k$$ most similar movies for each movie to the approximate user-item matrix. Here, we particularly focus on [DIMSUM](item_based_cf.html#dimsum-approximated-all-pairs-cosine-similarity-computation), @@ -364,12 +364,12 @@ Here, we predict ratng values of binarized user-item rating matrix of testing da Based on predicted rating scores, we can recommend top-k items for each user that he or she will be likely to put high scores. -## Predict unknown value of user-item matrix +## Predict unknown ratings of a user-item matrix -Based on known ratings and SLIM weight matrix, we can predict unknown values in the user-item matrix in `predicted`. +Based on known ratings and SLIM weight matrix, we predict unknown ratings in the user-item matrix. SLIM predicts ratings of user-item pairs based on top-$$K$$ similar items. -The `predict_pair` table represents candidates for recommended user-movie pairs, excluding known ratings in training dataset. +The `predict_pair` table represents candidates for recommended user-movie pairs, excluding known ratings in the training dataset. ```sql CREATE OR REPLACE VIEW predict_pair @@ -397,9 +397,9 @@ from where r.itemid IS NULL -- anti join ; -`` - ``` + +```sql -- optionally set the mean/default value of prediction set hivevar:mu=0.0; @@ -578,7 +578,6 @@ from | mrr | |:---:| | 0.10782647321821472 | -``` ### $$K$$-hold result