From 9d53fd47d2de164cf8a5527a753e66c8852a8f7f Mon Sep 17 00:00:00 2001
From: lthoang <trhoanglee@gmail.com>
Date: Sat, 30 Dec 2023 11:14:31 +0800
Subject: [PATCH 1/2] Fix spop scoring function
---
cornac/models/spop/recom_spop.py | 2 +-
examples/spop_yoochoose.py | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/cornac/models/spop/recom_spop.py b/cornac/models/spop/recom_spop.py
index 50f1795d..1910c607 100644
--- a/cornac/models/spop/recom_spop.py
+++ b/cornac/models/spop/recom_spop.py
@@ -48,7 +48,7 @@ def fit(self, train_set, val_set=None):
return self
def score(self, user_idx, history_items, **kwargs):
- item_scores = np.ones(self.total_items, dtype=np.float32)
+ item_scores = np.zeros(self.total_items, dtype=np.float32)
max_item_freq = max(self.item_freq.values()) if len(self.item_freq) > 0 else 1
for iid, freq in self.item_freq.items():
item_scores[iid] = freq / max_item_freq
diff --git a/examples/spop_yoochoose.py b/examples/spop_yoochoose.py
index ef87d9e2..eb4f785f 100644
--- a/examples/spop_yoochoose.py
+++ b/examples/spop_yoochoose.py
@@ -12,7 +12,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.
# ============================================================================
-"""Example of a next-basket recommendation model that merely uses item top frequency"""
+"""Example of a next-item recommendation model based on item popularity"""
import cornac
from cornac.datasets import yoochoose
From 1dc05a370a376c3b437961bd1ab1a39a6159b876 Mon Sep 17 00:00:00 2001
From: lthoang <trhoanglee@gmail.com>
Date: Sat, 30 Dec 2023 11:14:46 +0800
Subject: [PATCH 2/2] Update docs
---
README.md | 1 +
docs/source/api_ref/models.rst | 5 +++++
examples/README.md | 6 ++++++
3 files changed, 12 insertions(+)
diff --git a/README.md b/README.md
index 221cd804..bff1c841 100644
--- a/README.md
+++ b/README.md
@@ -196,6 +196,7 @@ The recommender models supported by Cornac are listed below. Why don't you join
| | [Most Popular (MostPop)](cornac/models/most_pop), [paper](https://arxiv.org/ftp/arxiv/papers/1205/1205.2618.pdf) | N/A | [bpr_netflix.py](examples/bpr_netflix.py)
| | [Non-negative Matrix Factorization (NMF)](cornac/models/nmf), [paper](http://papers.nips.cc/paper/1861-algorithms-for-non-negative-matrix-factorization.pdf) | N/A | [nmf_exp.py](examples/nmf_example.py)
| | [Probabilistic Matrix Factorization (PMF)](cornac/models/pmf), [paper](https://papers.nips.cc/paper/3208-probabilistic-matrix-factorization.pdf) | N/A | [pmf_ratio.py](examples/pmf_ratio.py)
+| | [Session Popular (SPop)](cornac/models/spop), [paper](https://arxiv.org/pdf/1511.06939.pdf) | N/A | [spop_yoochoose.py](examples/spop_yoochoose.py)
| | [Singular Value Decomposition (SVD)](cornac/models/svd), [paper](https://people.engr.tamu.edu/huangrh/Spring16/papers_course/matrix_factorization.pdf) | N/A | [svd_exp.py](examples/svd_example.py)
| | [Social Recommendation using PMF (SoRec)](cornac/models/sorec), [paper](http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.304.2464&rep=rep1&type=pdf) | N/A | [sorec_filmtrust.py](examples/sorec_filmtrust.py)
| | [User K-Nearest-Neighbors (UserKNN)](cornac/models/knn), [paper](https://arxiv.org/pdf/1301.7363.pdf) | N/A | [knn_movielens.py](examples/knn_movielens.py)
diff --git a/docs/source/api_ref/models.rst b/docs/source/api_ref/models.rst
index bbbc6a1a..91b760ef 100644
--- a/docs/source/api_ref/models.rst
+++ b/docs/source/api_ref/models.rst
@@ -259,6 +259,11 @@ Probabilitic Matrix Factorization (PMF)
.. automodule:: cornac.models.pmf.recom_pmf
:members:
+Session Popular (SPop)
+----------------------
+.. automodule:: cornac.models.spop.recom_spop
+ :members:
+
Singular Value Decomposition (SVD)
----------------------------------
.. automodule:: cornac.models.svd.recom_svd
diff --git a/examples/README.md b/examples/README.md
index e40c9e3e..c5d298ae 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -108,6 +108,12 @@
----
+## Next-Item Algorithms
+
+[spop_yoochoose.py](spop_yoochoose.py) - Next-item recommendation based on item popularity.
+
+----
+
## Next-Basket Algorithms
[gp_top_tafeng.py](gp_top_tafeng.py) - Next-basket recommendation model that merely uses item top frequency.