SVD++ - Revision history

Zeno Gantner at 19:16, 30 May 2012

2012-05-30T19:16:43Z

2012-05-30T15:50:30Z

Model Learning

2012-02-05T22:33:32Z

Implementations

2012-02-05T17:16:39Z

2012-02-05T17:16:03Z

2012-02-05T16:20:31Z

Implementations

2012-02-05T16:11:48Z

Implementations

2011-11-12T09:57:03Z

2011-11-12T09:56:39Z

Implementations

2011-10-04T06:16:24Z

@@ Line 18: / Line 18: @@
 with the user feedback information. With the most two common choices: (1) <math>\frac{1}{\sqrt{|N(u)|}}</math> for implicit feedback, (2) <math>\frac{r_{uj} - b_u}{\sqrt{|R(u)|}}</math> for explicit feedback.
-== Model Learning ==
+== Learning ==
-* SVD++ can be trained using [[ALS]].
+SVD++ can be trained using [[ALS]].
-== Efficient SGD Training for SVD++ ==
+It is slow to train a SVD++-style model using [[stochastic gradient descent]] due to the size of user feedback information, however, an efficient SGD training algorithm can be used.
-http://arxiv.org/abs/1109.2271 describes efficient training with user feedback information in section 4
+[http://arxiv.org/abs/1109.2271] describes efficient training with user feedback information in section 4
 == Literature ==
@@ Line 33: / Line 32: @@
 * [[LibFM]] can also be used to implement SVD++
 * [http://gustavonarea.net/blog/posts/korens-svd-python-implementation/ wooflix] is a (not very fast) [[Python]] implementation of SVD++
-* [[MyMediaLite]]: [https://github.com/zenogantner/MyMediaLite/blob/master/src/MyMediaLite/RatingPrediction/SVDPlusPlus.cs SVD++ source code on GitHub]
+* [[MyMediaLite]]: [https://github.com/zenogantner/MyMediaLite/blob/master/src/MyMediaLite/RatingPrediction/SVDPlusPlus.cs SVD++ source code on GitHub]; see also [https://github.com/zenogantner/MyMediaLite/blob/master/src/MyMediaLite/RatingPrediction/SigmoidSVDPlusPlus.cs]
 [[Category:Method]]

@@ Line 19: / Line 19: @@
 == Model Learning ==
-* SVD++ can be trained using ALS.
+* SVD++ can be trained using [[ALS]].
 * It's slow to train a SVD++ style model using stochastic gradient descent due to the size of user feedback information, however, an efficient SGD training algorithm can be used.

@@ Line 33: / Line 33: @@
 * [[LibFM]] can also be used to implement SVD++
 * [http://gustavonarea.net/blog/posts/korens-svd-python-implementation/ wooflix] is a (not very fast) [[Python]] implementation of SVD++
 [[Category:Method]]

@@ Line 5: / Line 5: @@
 The SVD++ model is formally described as following equation:
-<math>r_{ui} = \mu + b_u + b_i + \left(p_u + \frac{1}{\sqrt{|N(u)|}}\sum_{j\in N(u)} y_j \right)^T q_i</math>
+<math>r_{ui} = \mu + b_u + b_i + \left(p_u + \frac{1}{\sqrt{|N(u)|}}\sum_{j\in N(u)} y_j \right)^T q_i,</math>
-Where <math>N(u)</math> is the set of implicit information( the set of items user u rated ).
+where <math>N(u)</math> is the set of implicit information( the set of items user u rated ).
 == General Formalization for User Feedback Information ==
 A more general form of utilizing implicit/explicit information as user factor can be described in following equation
-<math>r_{ui} = \mu + b_u + b_i + \left(p_u + \sum_{i\in Ufeed(u)} \alpha_i y_i \right)^T q_i</math>
+<math>r_{ui} = \mu + b_u + b_i + \left(p_u + \sum_{i\in Ufeed(u)} \alpha_i y_i \right)^T q_i.</math>
 Here <math>Ufeed(u)</math> is the set of user feedback information( e.g: the web pages the user clicked, the music on users' favorite list,

@@ Line 1: / Line 1: @@
-'''SVD++''' refers to the matrix factorization algorithm which makes use of [[implicit feedback]] information.
+'''SVD++''' refers to a [[matrix factorization]] model which makes use of [[implicit feedback]] information.
-In general, implicit feedback can refer to any kinds of users' history information that can help indicate users'
+In general, implicit feedback can refer to any kinds of [[user]]s' history information that can help indicate users' [[preference]].
 == Model Formalization ==

@@ Line 32: / Line 32: @@
 * The [[GraphLab]] Collaborative Filtering Library has implemented SVD++ for multicore: http://graphlab.org/pmf.html
 * [[SVDFeature]] is a toolkit designed for feature-based matrix factorization, can be used to implement SVD++ and its extensions.
 [[Category:Method]]

@@ Line 30: / Line 30: @@
 == Implementations ==
-* GraphLab Collaborative Filtering Library has implemented SVD++ for multicore: http://graphlab.org/pmf.html
+* The [[GraphLab]] Collaborative Filtering Library has implemented SVD++ for multicore: http://graphlab.org/pmf.html
-* [[SVDFeature]] is a toolkit designed for feature-based matrix factorization, can be used to implement SVD++ and it's extensions.
+* [[SVDFeature]] is a toolkit designed for feature-based matrix factorization, can be used to implement SVD++ and its extensions.
 [[Category:Method]]

@@ Line 27: / Line 27: @@
 == Literature ==
-* [[Yehuda Koren]]: Factorization meets the neighborhood: a multifaceted collaborative filtering model, KDD 2008,http://portal.acm.org/citation.cfm?id=1401890.1401944
+* [[Yehuda Koren]]: Factorization meets the neighborhood: a multifaceted collaborative filtering model, KDD 2008, http://portal.acm.org/citation.cfm?id=1401890.1401944
 == Implementations ==