SlideShare a Scribd company logo
1 of 15
Fifty Shades of Ratings:
How to Benefit from a Negative Feedback in
Top-N Recommendations Tasks
by Evgeny Frolov1 and Ivan Oseledets1, 2
1Skolkovo Institute of Science and Technology
2Institute of Numerical Mathematics of the Russian Academy of Sciences
“Almost” cold-start problem
Recommendations are insensitive to negative “signal”.
Shift of recommendations paradigm:
Is this a good list of recommendations?
new user
Users may share not only what they love, but also what they hate.
Why standard approach fails?
𝒑 𝑇
𝒒 𝑇new user row
𝐴 ≈ 𝑈
Σ 𝑉 𝑇
Pure SVD* of matrix of ratings 𝐴users
movies
*P. Cremonesi, Y.Koren, R.Turrin, "Performance of Recommender Algorithms on Top-N Recommendation Tasks“, 2010
𝒓 ≈ 𝑉𝑉 𝑇
𝒑
vector of predicted item scores
approximate update to SVD generated by 𝒑
toprec 𝒑, 𝑛 ≔ arg max 𝒓
𝑛
top-𝒏 recommendations task
𝒓 𝑇 = 𝒒 𝑇Σ𝑉 𝑇 ≈ 𝒑 𝑇 𝑉Σ−1Σ𝑉 𝑇 = 𝒑 𝑇 𝑉𝑉 𝑇folding-in:
arg max 𝑉𝑉 𝑇
0, … , 0, 𝟐, 0, … , 0 𝑇
≡ arg max 𝑉𝑉 𝑇
0, … , 0, 𝟓, 0, … , 0 𝑇
How to solve this problem?
Rating elicitation hard to peak most representative items
increases barrier to entry (not effortless for user)
non-personalized user experience
Typical approach:
meaningful recommendations even from a single feedback
respect feedback polarity
no heuristics, no side information
generalize well on other scenarios (not only cold-start)
Requirements:
Technique: Matrix factorization
Restating the problem
𝑈𝑠𝑒𝑟 × 𝐼𝑡𝑒𝑚 → 𝑅𝑎𝑡𝑖𝑛𝑔
Users
Items
3
Standard model
Users
3
1 2
54
1
* T. G. Kolda and B. W. Bader, “Tensor Decompositions and Applications”, 2009
𝑈𝑠𝑒𝑟 × 𝐼𝑡𝑒𝑚 × 𝑅𝑎𝑡𝑖𝑛𝑔 → 𝑅𝑒𝑙𝑒𝑣𝑎𝑛𝑐𝑒 𝑆𝑐𝑜𝑟𝑒
Collaborative Full Feedback model
CoFFee (proposed approach)
Technique: Tensor Factorization
based on Tucker Decomposition*
𝒜 ≈ 𝒢 ×1 𝑈 ×2 𝑉 ×3 𝑊
ratings are cardinal values
Recommendations in real-time
𝑃 – matrix of new
user preferences approximate row update𝒒 𝑻
𝑅 ≈ 𝑉𝑉 𝑇 𝑃𝑊𝑊 𝑇 items relevance matrix
Compare to SVD: 𝒓 ≈ 𝑉𝑉 𝑇
𝒑
𝒢
𝑈
𝑊
𝑉
𝒜 ≈
Users 𝒜 ≈ 𝒢 ×1 𝑈 ×2 𝑉 ×3 𝑊
Higher order folding-in: “Shades of ratings”
𝑊 embeds ratings onto
latent feature space!
“Shades” of ratings
Model is equally sensitive
to any kind of feedback.
Granular view of user preferences,
concerning all possible ratings.
More dense colors correspond to higher relevance score.
ratings
movies
1 2 3 4 50
rankingtask
𝑅 ≈ 𝑉𝑉 𝑇 𝑃𝑊𝑊 𝑇
rating prediction
Cold-start with CoFFee
Undesired positivity bias in evaluation
Precision =
1
#(test users)
test
users
#(recommended items ∩ holdout items)
#(recommended items)
𝐷𝐶𝐺 =
𝑖
2 𝑟𝑒𝑙 𝑖 − 1
log2(𝑖 + 1)
Need to distinguish between relevant and irrelevant recommendations
Implicit assumption: all recommendations are interesting to the user.
𝑟𝑒𝑙𝑖 - true rating of a recommended item at position 𝑖
Low ratings do not express enjoyment!
Redefining metrics
2 3 4 5
+ + +
𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 =
𝑡𝑝
𝑡𝑝 + 𝑓𝑝
𝑅𝑒𝑐𝑎𝑙𝑙 =
𝑡𝑝
𝑡𝑝 + 𝑓𝑛
Relevance based
Ranking based 𝐷𝐶𝐺 =
𝑝
2 𝑟 𝑝 − 1
log2(𝑝 + 1) 𝑝 ∶ {𝑟𝑝 ≥ positivity threshold}
𝑟𝑝 - value of positive feedback
New metric
Discounted Cumulative Loss
𝐷𝐶𝐿 =
𝑛
2−𝑟 𝑛 − 1
−log2(𝑛 + 1) 𝑛: {0 < 𝑟𝑛 < positivity threshold}
𝑟𝑛 - value of negative feedback
Holdout items
Recommendations
tpfptn fn
“presumption of innocence”
“Almost” cold-start with 1 negative feedback
Data: Movielens 10M
Recommendations for “known user”
Data: Movielens 10M
Key takeaways
Standard evaluation metrics are biased towards positive effects of recommendations.
Negative feedback is a valuable source of information and shouldn’t be neglected.
It’s more natural to treat users’ feedback as ordinal not cardinal concept.
Tensor methods are effective for this kind of problems, giving you speed and quality.
Proposed CoFFee model can help to alleviate rating elicitation problems.
Polara framework
fast and easy-to-use
feature-rich and extensible
actively developed
MyMediaLite support (extended with folding-in)
https://github.com/evfro/polara
“RecSys for Humans”
Questions?
evgeny.frolov@skoltech.ru i.oseledets@skoltech.ru
Fixed-state code to reproduce results
https://github.com/evfro/fifty-shades
Run it right in your browser!
http://mybinder.org/repo/evfro/fifty-shades

More Related Content

What's hot

VBAで数値計算 10 逆行列と疑似逆行列
VBAで数値計算 10 逆行列と疑似逆行列VBAで数値計算 10 逆行列と疑似逆行列
VBAで数値計算 10 逆行列と疑似逆行列Katsuhiro Morishita
 
東京都市大学 データ解析入門 4 スパース性と圧縮センシング1
東京都市大学 データ解析入門 4 スパース性と圧縮センシング1東京都市大学 データ解析入門 4 スパース性と圧縮センシング1
東京都市大学 データ解析入門 4 スパース性と圧縮センシング1hirokazutanaka
 
スパースモデリング入門
スパースモデリング入門スパースモデリング入門
スパースモデリング入門Hideo Terada
 
電路學 - [第四章] 儲能元件
電路學 - [第四章] 儲能元件電路學 - [第四章] 儲能元件
電路學 - [第四章] 儲能元件Simen Li
 
Introduction to Topological Data Analysis
Introduction to Topological Data AnalysisIntroduction to Topological Data Analysis
Introduction to Topological Data AnalysisMason Porter
 
[DL輪読会]Deep Learning 第16章 深層学習のための構造化確率モデル
[DL輪読会]Deep Learning 第16章 深層学習のための構造化確率モデル[DL輪読会]Deep Learning 第16章 深層学習のための構造化確率モデル
[DL輪読会]Deep Learning 第16章 深層学習のための構造化確率モデルDeep Learning JP
 
データ解析7 主成分分析の基礎
データ解析7 主成分分析の基礎データ解析7 主成分分析の基礎
データ解析7 主成分分析の基礎Hirotaka Hachiya
 
PRML輪読#12
PRML輪読#12PRML輪読#12
PRML輪読#12matsuolab
 
混合モデルとEMアルゴリズム(PRML第9章)
混合モデルとEMアルゴリズム(PRML第9章)混合モデルとEMアルゴリズム(PRML第9章)
混合モデルとEMアルゴリズム(PRML第9章)Takao Yamanaka
 
パーセプトロン型学習規則
パーセプトロン型学習規則パーセプトロン型学習規則
パーセプトロン型学習規則Shuhei Sowa
 
[DL輪読会]Deep Learning 第14章 自己符号化器
[DL輪読会]Deep Learning 第14章 自己符号化器[DL輪読会]Deep Learning 第14章 自己符号化器
[DL輪読会]Deep Learning 第14章 自己符号化器Deep Learning JP
 
[DL輪読会]Deep Learning 第13章 線形因子モデル
[DL輪読会]Deep Learning 第13章 線形因子モデル[DL輪読会]Deep Learning 第13章 線形因子モデル
[DL輪読会]Deep Learning 第13章 線形因子モデルDeep Learning JP
 
[PRML] パターン認識と機械学習(第3章:線形回帰モデル)
[PRML] パターン認識と機械学習(第3章:線形回帰モデル)[PRML] パターン認識と機械学習(第3章:線形回帰モデル)
[PRML] パターン認識と機械学習(第3章:線形回帰モデル)Ryosuke Sasaki
 
ぞくパタ最終回: 13章「共クラスタリング」
ぞくパタ最終回: 13章「共クラスタリング」ぞくパタ最終回: 13章「共クラスタリング」
ぞくパタ最終回: 13章「共クラスタリング」Akifumi Eguchi
 
Soluciones por series
Soluciones por seriesSoluciones por series
Soluciones por seriesKike Prieto
 
2022年度秋学期 画像情報処理 第9回 離散フーリエ変換と離散コサイン変換 (2022. 11. 25)
2022年度秋学期 画像情報処理 第9回 離散フーリエ変換と離散コサイン変換 (2022. 11. 25) 2022年度秋学期 画像情報処理 第9回 離散フーリエ変換と離散コサイン変換 (2022. 11. 25)
2022年度秋学期 画像情報処理 第9回 離散フーリエ変換と離散コサイン変換 (2022. 11. 25) Akira Asano
 

What's hot (20)

量子情報復習
量子情報復習量子情報復習
量子情報復習
 
VBAで数値計算 10 逆行列と疑似逆行列
VBAで数値計算 10 逆行列と疑似逆行列VBAで数値計算 10 逆行列と疑似逆行列
VBAで数値計算 10 逆行列と疑似逆行列
 
東京都市大学 データ解析入門 4 スパース性と圧縮センシング1
東京都市大学 データ解析入門 4 スパース性と圧縮センシング1東京都市大学 データ解析入門 4 スパース性と圧縮センシング1
東京都市大学 データ解析入門 4 スパース性と圧縮センシング1
 
Principal component analysis
Principal component analysisPrincipal component analysis
Principal component analysis
 
スパースモデリング入門
スパースモデリング入門スパースモデリング入門
スパースモデリング入門
 
電路學 - [第四章] 儲能元件
電路學 - [第四章] 儲能元件電路學 - [第四章] 儲能元件
電路學 - [第四章] 儲能元件
 
PRML chapter7
PRML chapter7PRML chapter7
PRML chapter7
 
Introduction to Topological Data Analysis
Introduction to Topological Data AnalysisIntroduction to Topological Data Analysis
Introduction to Topological Data Analysis
 
[DL輪読会]Deep Learning 第16章 深層学習のための構造化確率モデル
[DL輪読会]Deep Learning 第16章 深層学習のための構造化確率モデル[DL輪読会]Deep Learning 第16章 深層学習のための構造化確率モデル
[DL輪読会]Deep Learning 第16章 深層学習のための構造化確率モデル
 
データ解析7 主成分分析の基礎
データ解析7 主成分分析の基礎データ解析7 主成分分析の基礎
データ解析7 主成分分析の基礎
 
PRML輪読#12
PRML輪読#12PRML輪読#12
PRML輪読#12
 
混合モデルとEMアルゴリズム(PRML第9章)
混合モデルとEMアルゴリズム(PRML第9章)混合モデルとEMアルゴリズム(PRML第9章)
混合モデルとEMアルゴリズム(PRML第9章)
 
パーセプトロン型学習規則
パーセプトロン型学習規則パーセプトロン型学習規則
パーセプトロン型学習規則
 
[DL輪読会]Deep Learning 第14章 自己符号化器
[DL輪読会]Deep Learning 第14章 自己符号化器[DL輪読会]Deep Learning 第14章 自己符号化器
[DL輪読会]Deep Learning 第14章 自己符号化器
 
[DL輪読会]Deep Learning 第13章 線形因子モデル
[DL輪読会]Deep Learning 第13章 線形因子モデル[DL輪読会]Deep Learning 第13章 線形因子モデル
[DL輪読会]Deep Learning 第13章 線形因子モデル
 
[PRML] パターン認識と機械学習(第3章:線形回帰モデル)
[PRML] パターン認識と機械学習(第3章:線形回帰モデル)[PRML] パターン認識と機械学習(第3章:線形回帰モデル)
[PRML] パターン認識と機械学習(第3章:線形回帰モデル)
 
ぞくパタ最終回: 13章「共クラスタリング」
ぞくパタ最終回: 13章「共クラスタリング」ぞくパタ最終回: 13章「共クラスタリング」
ぞくパタ最終回: 13章「共クラスタリング」
 
Soluciones por series
Soluciones por seriesSoluciones por series
Soluciones por series
 
2022年度秋学期 画像情報処理 第9回 離散フーリエ変換と離散コサイン変換 (2022. 11. 25)
2022年度秋学期 画像情報処理 第9回 離散フーリエ変換と離散コサイン変換 (2022. 11. 25) 2022年度秋学期 画像情報処理 第9回 離散フーリエ変換と離散コサイン変換 (2022. 11. 25)
2022年度秋学期 画像情報処理 第9回 離散フーリエ変換と離散コサイン変換 (2022. 11. 25)
 
Prml07
Prml07Prml07
Prml07
 

Viewers also liked

(Some) pitfalls of distributed learning
(Some) pitfalls of distributed learning(Some) pitfalls of distributed learning
(Some) pitfalls of distributed learningYves Raimond
 
Balancing Discovery and Continuation in Recommendations
Balancing Discovery and Continuation in RecommendationsBalancing Discovery and Continuation in Recommendations
Balancing Discovery and Continuation in RecommendationsMohammad Hossein Taghavi
 
Contrasting Offline and Online Results when Evaluating Recommendation Algorithms
Contrasting Offline and Online Results when Evaluating Recommendation AlgorithmsContrasting Offline and Online Results when Evaluating Recommendation Algorithms
Contrasting Offline and Online Results when Evaluating Recommendation AlgorithmsMarco Rossetti
 
Machine Learning at Netflix Scale
Machine Learning at Netflix ScaleMachine Learning at Netflix Scale
Machine Learning at Netflix ScaleAish Fenton
 
Recsys 2016: Modeling Contextual Information in Session-Aware Recommender Sys...
Recsys 2016: Modeling Contextual Information in Session-Aware Recommender Sys...Recsys 2016: Modeling Contextual Information in Session-Aware Recommender Sys...
Recsys 2016: Modeling Contextual Information in Session-Aware Recommender Sys...Bartlomiej Twardowski
 
Balancing Discovery and continuation in recommendation (hossein taghavi netflix)
Balancing Discovery and continuation in recommendation (hossein taghavi netflix)Balancing Discovery and continuation in recommendation (hossein taghavi netflix)
Balancing Discovery and continuation in recommendation (hossein taghavi netflix)IntoTheMinds
 
RecSys 2016 Talk: Feature Selection For Human Recommenders
RecSys 2016 Talk: Feature Selection For Human RecommendersRecSys 2016 Talk: Feature Selection For Human Recommenders
RecSys 2016 Talk: Feature Selection For Human RecommendersKatherine Livins
 
Netflix Recommendations Using Spark + Cassandra (Prasanna Padmanabhan & Roopa...
Netflix Recommendations Using Spark + Cassandra (Prasanna Padmanabhan & Roopa...Netflix Recommendations Using Spark + Cassandra (Prasanna Padmanabhan & Roopa...
Netflix Recommendations Using Spark + Cassandra (Prasanna Padmanabhan & Roopa...DataStax
 
Past, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry PerspectivePast, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry PerspectiveJustin Basilico
 
Factorization Meets the Item Embedding: Regularizing Matrix Factorization wit...
Factorization Meets the Item Embedding: Regularizing Matrix Factorization wit...Factorization Meets the Item Embedding: Regularizing Matrix Factorization wit...
Factorization Meets the Item Embedding: Regularizing Matrix Factorization wit...Dawen Liang
 
Netflix's Recommendation ML Pipeline Using Apache Spark: Spark Summit East ta...
Netflix's Recommendation ML Pipeline Using Apache Spark: Spark Summit East ta...Netflix's Recommendation ML Pipeline Using Apache Spark: Spark Summit East ta...
Netflix's Recommendation ML Pipeline Using Apache Spark: Spark Summit East ta...Spark Summit
 

Viewers also liked (11)

(Some) pitfalls of distributed learning
(Some) pitfalls of distributed learning(Some) pitfalls of distributed learning
(Some) pitfalls of distributed learning
 
Balancing Discovery and Continuation in Recommendations
Balancing Discovery and Continuation in RecommendationsBalancing Discovery and Continuation in Recommendations
Balancing Discovery and Continuation in Recommendations
 
Contrasting Offline and Online Results when Evaluating Recommendation Algorithms
Contrasting Offline and Online Results when Evaluating Recommendation AlgorithmsContrasting Offline and Online Results when Evaluating Recommendation Algorithms
Contrasting Offline and Online Results when Evaluating Recommendation Algorithms
 
Machine Learning at Netflix Scale
Machine Learning at Netflix ScaleMachine Learning at Netflix Scale
Machine Learning at Netflix Scale
 
Recsys 2016: Modeling Contextual Information in Session-Aware Recommender Sys...
Recsys 2016: Modeling Contextual Information in Session-Aware Recommender Sys...Recsys 2016: Modeling Contextual Information in Session-Aware Recommender Sys...
Recsys 2016: Modeling Contextual Information in Session-Aware Recommender Sys...
 
Balancing Discovery and continuation in recommendation (hossein taghavi netflix)
Balancing Discovery and continuation in recommendation (hossein taghavi netflix)Balancing Discovery and continuation in recommendation (hossein taghavi netflix)
Balancing Discovery and continuation in recommendation (hossein taghavi netflix)
 
RecSys 2016 Talk: Feature Selection For Human Recommenders
RecSys 2016 Talk: Feature Selection For Human RecommendersRecSys 2016 Talk: Feature Selection For Human Recommenders
RecSys 2016 Talk: Feature Selection For Human Recommenders
 
Netflix Recommendations Using Spark + Cassandra (Prasanna Padmanabhan & Roopa...
Netflix Recommendations Using Spark + Cassandra (Prasanna Padmanabhan & Roopa...Netflix Recommendations Using Spark + Cassandra (Prasanna Padmanabhan & Roopa...
Netflix Recommendations Using Spark + Cassandra (Prasanna Padmanabhan & Roopa...
 
Past, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry PerspectivePast, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry Perspective
 
Factorization Meets the Item Embedding: Regularizing Matrix Factorization wit...
Factorization Meets the Item Embedding: Regularizing Matrix Factorization wit...Factorization Meets the Item Embedding: Regularizing Matrix Factorization wit...
Factorization Meets the Item Embedding: Regularizing Matrix Factorization wit...
 
Netflix's Recommendation ML Pipeline Using Apache Spark: Spark Summit East ta...
Netflix's Recommendation ML Pipeline Using Apache Spark: Spark Summit East ta...Netflix's Recommendation ML Pipeline Using Apache Spark: Spark Summit East ta...
Netflix's Recommendation ML Pipeline Using Apache Spark: Spark Summit East ta...
 

Similar to Fifty Shades of Ratings: How to Benefit from a Negative Feedback in Top-N Recommendations Tasks

Low rank models for recommender systems with limited preference information
Low rank models for recommender systems with limited preference informationLow rank models for recommender systems with limited preference information
Low rank models for recommender systems with limited preference informationEvgeny Frolov
 
A new similarity measurement based on hellinger distance for collaborating fi...
A new similarity measurement based on hellinger distance for collaborating fi...A new similarity measurement based on hellinger distance for collaborating fi...
A new similarity measurement based on hellinger distance for collaborating fi...Prabhu Kumar
 
Recommender Systems: Advances in Collaborative Filtering
Recommender Systems: Advances in Collaborative FilteringRecommender Systems: Advances in Collaborative Filtering
Recommender Systems: Advances in Collaborative FilteringChangsung Moon
 
Empirical Evaluation of Active Learning in Recommender Systems
Empirical Evaluation of Active Learning in Recommender SystemsEmpirical Evaluation of Active Learning in Recommender Systems
Empirical Evaluation of Active Learning in Recommender SystemsUniversity of Bergen
 
Recommender system
Recommender systemRecommender system
Recommender systemYinghan Fu
 
ACM ICTIR 2019 Slides - Santa Clara, USA
ACM ICTIR 2019 Slides -  Santa Clara, USAACM ICTIR 2019 Slides -  Santa Clara, USA
ACM ICTIR 2019 Slides - Santa Clara, USAIadh Ounis
 
Review: [KDD'21]Model-Agnostic Counterfactual Reasoning for Eliminating Popul...
Review: [KDD'21]Model-Agnostic Counterfactual Reasoning for Eliminating Popul...Review: [KDD'21]Model-Agnostic Counterfactual Reasoning for Eliminating Popul...
Review: [KDD'21]Model-Agnostic Counterfactual Reasoning for Eliminating Popul...CS Kwak
 
[ICDM'22] Mitigating Popularity Bias in Recommendation with Unbalanced Intera...
[ICDM'22] Mitigating Popularity Bias in Recommendation with Unbalanced Intera...[ICDM'22] Mitigating Popularity Bias in Recommendation with Unbalanced Intera...
[ICDM'22] Mitigating Popularity Bias in Recommendation with Unbalanced Intera...CS Kwak
 
(141205) Masters_Thesis_Defense_Sundong_Kim
(141205) Masters_Thesis_Defense_Sundong_Kim(141205) Masters_Thesis_Defense_Sundong_Kim
(141205) Masters_Thesis_Defense_Sundong_KimSundong Kim
 
GRU4Rec v2 - Recurrent Neural Networks with Top-k Gains for Session-based Rec...
GRU4Rec v2 - Recurrent Neural Networks with Top-k Gains for Session-based Rec...GRU4Rec v2 - Recurrent Neural Networks with Top-k Gains for Session-based Rec...
GRU4Rec v2 - Recurrent Neural Networks with Top-k Gains for Session-based Rec...Balázs Hidasi
 
A Scalable, High-performance Algorithm for Hybrid Job Recommendations
A Scalable, High-performance Algorithm for Hybrid Job RecommendationsA Scalable, High-performance Algorithm for Hybrid Job Recommendations
A Scalable, High-performance Algorithm for Hybrid Job RecommendationsToon De Pessemier
 
Rokach-GomaxSlides.pptx
Rokach-GomaxSlides.pptxRokach-GomaxSlides.pptx
Rokach-GomaxSlides.pptxJadna Almeida
 
Rokach-GomaxSlides (1).pptx
Rokach-GomaxSlides (1).pptxRokach-GomaxSlides (1).pptx
Rokach-GomaxSlides (1).pptxJadna Almeida
 
Recommender Systems Fairness Evaluation via Generalized Cross Entropy
Recommender Systems Fairness Evaluation via Generalized Cross EntropyRecommender Systems Fairness Evaluation via Generalized Cross Entropy
Recommender Systems Fairness Evaluation via Generalized Cross EntropyVito Walter Anelli
 
What recommender systems can learn from decision psychology about preference ...
What recommender systems can learn from decision psychology about preference ...What recommender systems can learn from decision psychology about preference ...
What recommender systems can learn from decision psychology about preference ...Eindhoven University of Technology / JADS
 
Download
DownloadDownload
Downloadbutest
 
Download
DownloadDownload
Downloadbutest
 
The Wisdom of the Few @SIGIR09
The Wisdom of the Few @SIGIR09The Wisdom of the Few @SIGIR09
The Wisdom of the Few @SIGIR09Xavier Amatriain
 
Recommender Systems from A to Z – Model Evaluation
Recommender Systems from A to Z – Model EvaluationRecommender Systems from A to Z – Model Evaluation
Recommender Systems from A to Z – Model EvaluationCrossing Minds
 

Similar to Fifty Shades of Ratings: How to Benefit from a Negative Feedback in Top-N Recommendations Tasks (20)

Low rank models for recommender systems with limited preference information
Low rank models for recommender systems with limited preference informationLow rank models for recommender systems with limited preference information
Low rank models for recommender systems with limited preference information
 
A new similarity measurement based on hellinger distance for collaborating fi...
A new similarity measurement based on hellinger distance for collaborating fi...A new similarity measurement based on hellinger distance for collaborating fi...
A new similarity measurement based on hellinger distance for collaborating fi...
 
Recommender Systems: Advances in Collaborative Filtering
Recommender Systems: Advances in Collaborative FilteringRecommender Systems: Advances in Collaborative Filtering
Recommender Systems: Advances in Collaborative Filtering
 
Empirical Evaluation of Active Learning in Recommender Systems
Empirical Evaluation of Active Learning in Recommender SystemsEmpirical Evaluation of Active Learning in Recommender Systems
Empirical Evaluation of Active Learning in Recommender Systems
 
Recommender system
Recommender systemRecommender system
Recommender system
 
ACM ICTIR 2019 Slides - Santa Clara, USA
ACM ICTIR 2019 Slides -  Santa Clara, USAACM ICTIR 2019 Slides -  Santa Clara, USA
ACM ICTIR 2019 Slides - Santa Clara, USA
 
Review: [KDD'21]Model-Agnostic Counterfactual Reasoning for Eliminating Popul...
Review: [KDD'21]Model-Agnostic Counterfactual Reasoning for Eliminating Popul...Review: [KDD'21]Model-Agnostic Counterfactual Reasoning for Eliminating Popul...
Review: [KDD'21]Model-Agnostic Counterfactual Reasoning for Eliminating Popul...
 
[ICDM'22] Mitigating Popularity Bias in Recommendation with Unbalanced Intera...
[ICDM'22] Mitigating Popularity Bias in Recommendation with Unbalanced Intera...[ICDM'22] Mitigating Popularity Bias in Recommendation with Unbalanced Intera...
[ICDM'22] Mitigating Popularity Bias in Recommendation with Unbalanced Intera...
 
(141205) Masters_Thesis_Defense_Sundong_Kim
(141205) Masters_Thesis_Defense_Sundong_Kim(141205) Masters_Thesis_Defense_Sundong_Kim
(141205) Masters_Thesis_Defense_Sundong_Kim
 
GRU4Rec v2 - Recurrent Neural Networks with Top-k Gains for Session-based Rec...
GRU4Rec v2 - Recurrent Neural Networks with Top-k Gains for Session-based Rec...GRU4Rec v2 - Recurrent Neural Networks with Top-k Gains for Session-based Rec...
GRU4Rec v2 - Recurrent Neural Networks with Top-k Gains for Session-based Rec...
 
A Scalable, High-performance Algorithm for Hybrid Job Recommendations
A Scalable, High-performance Algorithm for Hybrid Job RecommendationsA Scalable, High-performance Algorithm for Hybrid Job Recommendations
A Scalable, High-performance Algorithm for Hybrid Job Recommendations
 
Rokach-GomaxSlides.pptx
Rokach-GomaxSlides.pptxRokach-GomaxSlides.pptx
Rokach-GomaxSlides.pptx
 
Rokach-GomaxSlides (1).pptx
Rokach-GomaxSlides (1).pptxRokach-GomaxSlides (1).pptx
Rokach-GomaxSlides (1).pptx
 
Recommender Systems Fairness Evaluation via Generalized Cross Entropy
Recommender Systems Fairness Evaluation via Generalized Cross EntropyRecommender Systems Fairness Evaluation via Generalized Cross Entropy
Recommender Systems Fairness Evaluation via Generalized Cross Entropy
 
What recommender systems can learn from decision psychology about preference ...
What recommender systems can learn from decision psychology about preference ...What recommender systems can learn from decision psychology about preference ...
What recommender systems can learn from decision psychology about preference ...
 
Download
DownloadDownload
Download
 
Download
DownloadDownload
Download
 
The Wisdom of the Few @SIGIR09
The Wisdom of the Few @SIGIR09The Wisdom of the Few @SIGIR09
The Wisdom of the Few @SIGIR09
 
Recommender Systems from A to Z – Model Evaluation
Recommender Systems from A to Z – Model EvaluationRecommender Systems from A to Z – Model Evaluation
Recommender Systems from A to Z – Model Evaluation
 
DMAIC
DMAICDMAIC
DMAIC
 

Recently uploaded

Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% SecureCall me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% SecurePooja Nehwal
 
Halmar dropshipping via API with DroFx
Halmar  dropshipping  via API with DroFxHalmar  dropshipping  via API with DroFx
Halmar dropshipping via API with DroFxolyaivanovalion
 
RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998YohFuh
 
Introduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxIntroduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxfirstjob4
 
100-Concepts-of-AI by Anupama Kate .pptx
100-Concepts-of-AI by Anupama Kate .pptx100-Concepts-of-AI by Anupama Kate .pptx
100-Concepts-of-AI by Anupama Kate .pptxAnupama Kate
 
Mature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxMature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxolyaivanovalion
 
April 2024 - Crypto Market Report's Analysis
April 2024 - Crypto Market Report's AnalysisApril 2024 - Crypto Market Report's Analysis
April 2024 - Crypto Market Report's Analysismanisha194592
 
BigBuy dropshipping via API with DroFx.pptx
BigBuy dropshipping via API with DroFx.pptxBigBuy dropshipping via API with DroFx.pptx
BigBuy dropshipping via API with DroFx.pptxolyaivanovalion
 
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改atducpo
 
04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationshipsccctableauusergroup
 
VidaXL dropshipping via API with DroFx.pptx
VidaXL dropshipping via API with DroFx.pptxVidaXL dropshipping via API with DroFx.pptx
VidaXL dropshipping via API with DroFx.pptxolyaivanovalion
 
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Serviceranjana rawat
 
Generative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusGenerative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusTimothy Spann
 
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Callshivangimorya083
 
B2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxB2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxStephen266013
 
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiLow Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiSuhani Kapoor
 
Data-Analysis for Chicago Crime Data 2023
Data-Analysis for Chicago Crime Data  2023Data-Analysis for Chicago Crime Data  2023
Data-Analysis for Chicago Crime Data 2023ymrp368
 

Recently uploaded (20)

Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% SecureCall me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
 
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Kishangarh
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in  KishangarhDelhi 99530 vip 56974 Genuine Escort Service Call Girls in  Kishangarh
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Kishangarh
 
Halmar dropshipping via API with DroFx
Halmar  dropshipping  via API with DroFxHalmar  dropshipping  via API with DroFx
Halmar dropshipping via API with DroFx
 
RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998
 
Introduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxIntroduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptx
 
100-Concepts-of-AI by Anupama Kate .pptx
100-Concepts-of-AI by Anupama Kate .pptx100-Concepts-of-AI by Anupama Kate .pptx
100-Concepts-of-AI by Anupama Kate .pptx
 
Mature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxMature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptx
 
April 2024 - Crypto Market Report's Analysis
April 2024 - Crypto Market Report's AnalysisApril 2024 - Crypto Market Report's Analysis
April 2024 - Crypto Market Report's Analysis
 
BigBuy dropshipping via API with DroFx.pptx
BigBuy dropshipping via API with DroFx.pptxBigBuy dropshipping via API with DroFx.pptx
BigBuy dropshipping via API with DroFx.pptx
 
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
 
04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships
 
Sampling (random) method and Non random.ppt
Sampling (random) method and Non random.pptSampling (random) method and Non random.ppt
Sampling (random) method and Non random.ppt
 
VidaXL dropshipping via API with DroFx.pptx
VidaXL dropshipping via API with DroFx.pptxVidaXL dropshipping via API with DroFx.pptx
VidaXL dropshipping via API with DroFx.pptx
 
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
 
Generative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusGenerative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and Milvus
 
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
 
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
 
B2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxB2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docx
 
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiLow Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
 
Data-Analysis for Chicago Crime Data 2023
Data-Analysis for Chicago Crime Data  2023Data-Analysis for Chicago Crime Data  2023
Data-Analysis for Chicago Crime Data 2023
 

Fifty Shades of Ratings: How to Benefit from a Negative Feedback in Top-N Recommendations Tasks

  • 1. Fifty Shades of Ratings: How to Benefit from a Negative Feedback in Top-N Recommendations Tasks by Evgeny Frolov1 and Ivan Oseledets1, 2 1Skolkovo Institute of Science and Technology 2Institute of Numerical Mathematics of the Russian Academy of Sciences
  • 2. “Almost” cold-start problem Recommendations are insensitive to negative “signal”. Shift of recommendations paradigm: Is this a good list of recommendations? new user Users may share not only what they love, but also what they hate.
  • 3. Why standard approach fails? 𝒑 𝑇 𝒒 𝑇new user row 𝐴 ≈ 𝑈 Σ 𝑉 𝑇 Pure SVD* of matrix of ratings 𝐴users movies *P. Cremonesi, Y.Koren, R.Turrin, "Performance of Recommender Algorithms on Top-N Recommendation Tasks“, 2010 𝒓 ≈ 𝑉𝑉 𝑇 𝒑 vector of predicted item scores approximate update to SVD generated by 𝒑 toprec 𝒑, 𝑛 ≔ arg max 𝒓 𝑛 top-𝒏 recommendations task 𝒓 𝑇 = 𝒒 𝑇Σ𝑉 𝑇 ≈ 𝒑 𝑇 𝑉Σ−1Σ𝑉 𝑇 = 𝒑 𝑇 𝑉𝑉 𝑇folding-in: arg max 𝑉𝑉 𝑇 0, … , 0, 𝟐, 0, … , 0 𝑇 ≡ arg max 𝑉𝑉 𝑇 0, … , 0, 𝟓, 0, … , 0 𝑇
  • 4. How to solve this problem? Rating elicitation hard to peak most representative items increases barrier to entry (not effortless for user) non-personalized user experience Typical approach: meaningful recommendations even from a single feedback respect feedback polarity no heuristics, no side information generalize well on other scenarios (not only cold-start) Requirements:
  • 5. Technique: Matrix factorization Restating the problem 𝑈𝑠𝑒𝑟 × 𝐼𝑡𝑒𝑚 → 𝑅𝑎𝑡𝑖𝑛𝑔 Users Items 3 Standard model Users 3 1 2 54 1 * T. G. Kolda and B. W. Bader, “Tensor Decompositions and Applications”, 2009 𝑈𝑠𝑒𝑟 × 𝐼𝑡𝑒𝑚 × 𝑅𝑎𝑡𝑖𝑛𝑔 → 𝑅𝑒𝑙𝑒𝑣𝑎𝑛𝑐𝑒 𝑆𝑐𝑜𝑟𝑒 Collaborative Full Feedback model CoFFee (proposed approach) Technique: Tensor Factorization based on Tucker Decomposition* 𝒜 ≈ 𝒢 ×1 𝑈 ×2 𝑉 ×3 𝑊 ratings are cardinal values
  • 6. Recommendations in real-time 𝑃 – matrix of new user preferences approximate row update𝒒 𝑻 𝑅 ≈ 𝑉𝑉 𝑇 𝑃𝑊𝑊 𝑇 items relevance matrix Compare to SVD: 𝒓 ≈ 𝑉𝑉 𝑇 𝒑 𝒢 𝑈 𝑊 𝑉 𝒜 ≈ Users 𝒜 ≈ 𝒢 ×1 𝑈 ×2 𝑉 ×3 𝑊 Higher order folding-in: “Shades of ratings” 𝑊 embeds ratings onto latent feature space!
  • 7. “Shades” of ratings Model is equally sensitive to any kind of feedback. Granular view of user preferences, concerning all possible ratings. More dense colors correspond to higher relevance score. ratings movies 1 2 3 4 50 rankingtask 𝑅 ≈ 𝑉𝑉 𝑇 𝑃𝑊𝑊 𝑇 rating prediction
  • 9. Undesired positivity bias in evaluation Precision = 1 #(test users) test users #(recommended items ∩ holdout items) #(recommended items) 𝐷𝐶𝐺 = 𝑖 2 𝑟𝑒𝑙 𝑖 − 1 log2(𝑖 + 1) Need to distinguish between relevant and irrelevant recommendations Implicit assumption: all recommendations are interesting to the user. 𝑟𝑒𝑙𝑖 - true rating of a recommended item at position 𝑖 Low ratings do not express enjoyment!
  • 10. Redefining metrics 2 3 4 5 + + + 𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 = 𝑡𝑝 𝑡𝑝 + 𝑓𝑝 𝑅𝑒𝑐𝑎𝑙𝑙 = 𝑡𝑝 𝑡𝑝 + 𝑓𝑛 Relevance based Ranking based 𝐷𝐶𝐺 = 𝑝 2 𝑟 𝑝 − 1 log2(𝑝 + 1) 𝑝 ∶ {𝑟𝑝 ≥ positivity threshold} 𝑟𝑝 - value of positive feedback New metric Discounted Cumulative Loss 𝐷𝐶𝐿 = 𝑛 2−𝑟 𝑛 − 1 −log2(𝑛 + 1) 𝑛: {0 < 𝑟𝑛 < positivity threshold} 𝑟𝑛 - value of negative feedback Holdout items Recommendations tpfptn fn “presumption of innocence”
  • 11. “Almost” cold-start with 1 negative feedback Data: Movielens 10M
  • 12. Recommendations for “known user” Data: Movielens 10M
  • 13. Key takeaways Standard evaluation metrics are biased towards positive effects of recommendations. Negative feedback is a valuable source of information and shouldn’t be neglected. It’s more natural to treat users’ feedback as ordinal not cardinal concept. Tensor methods are effective for this kind of problems, giving you speed and quality. Proposed CoFFee model can help to alleviate rating elicitation problems.
  • 14. Polara framework fast and easy-to-use feature-rich and extensible actively developed MyMediaLite support (extended with folding-in) https://github.com/evfro/polara “RecSys for Humans”
  • 15. Questions? evgeny.frolov@skoltech.ru i.oseledets@skoltech.ru Fixed-state code to reproduce results https://github.com/evfro/fifty-shades Run it right in your browser! http://mybinder.org/repo/evfro/fifty-shades

Editor's Notes

  1. general conclusion: many models are unable to properly handle polarity of user feedback without additional heuristics and manual tweaking
  2. Key idea: represent ratings as an additional (categorical) variable and encode observations as a multidimensional array (tensor): Each interaction can now be encoded with 3 indices instead of two, as we take rating information into account in addition to users and items. We will call this multidimensional array a tensor and we use efficient tensor-based techniques
  3. Calculation of tensor-based model might be time consuming and we propose an efficient way of fast recommendations computation based on an generalization of known folding-in technique to higher order Tucker Decomposition obtained with HOOI
  4. This uncovers new recommendation scenarios beyond “users who like this also like…”