client-side hybrid rating prediction for recommendation · client-side hybrid rating prediction for...
TRANSCRIPT
![Page 1: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/1.jpg)
Client-side hybrid rating prediction forrecommendation
Andres Moreno12 Harold Castro 1 Michel Riveill 2
1School of EngineeringUniversidad de los Andes, Bogota, Colombia
2I3SUniversite de Nice Sophia Antipolis, France
UMAP, 2014
![Page 2: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/2.jpg)
Outline
Motivation: Privacy in recommender systemsRecommender SystemsPrivacy considerations
A client-side agent for recommendationAppliying client-side predictive modelsContent-based modelCollaborative Filtering model (CF)Hybrid prediction under expert advice
Final considerations
![Page 3: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/3.jpg)
Outline
Motivation: Privacy in recommender systemsRecommender SystemsPrivacy considerations
A client-side agent for recommendationAppliying client-side predictive modelsContent-based modelCollaborative Filtering model (CF)Hybrid prediction under expert advice
Final considerations
![Page 4: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/4.jpg)
Recommender systems
I Recommender systems are personalization systems thatautomatically calculate the relevance of a large collection ofdata items for a user. The relevance mapping between usersand items is used to select, screen out or rank items based onher preferences and situation.
U1
f1 fk
Uu
log files
item profiles
I1
f1 fk
Ii
Recommendation component
Interaction log component
feedback
item suggestionsTraining component
Prediction component
user profiles
Recommendation server
![Page 5: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/5.jpg)
Recommender systems
I Recommender systems are personalization systems thatautomatically calculate the relevance of a large collection ofdata items for a user. The relevance mapping between usersand items is used to select, screen out or rank items based onher preferences and situation.
U1
f1 fk
Uu
log files
item profiles
I1
f1 fk
Ii
Recommendation component
Interaction log component
feedback
item suggestionsTraining component
Prediction component
user profiles
Recommendation server
![Page 6: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/6.jpg)
Outline
Motivation: Privacy in recommender systemsRecommender SystemsPrivacy considerations
A client-side agent for recommendationAppliying client-side predictive modelsContent-based modelCollaborative Filtering model (CF)Hybrid prediction under expert advice
Final considerations
![Page 7: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/7.jpg)
Privacy considerations
I Recommender systems gather information about users andstore it in a centralized entity, then they apply heuristics ordata mining techniques to learn the users’ interests with thepurpose of detecting which elements are relevant for the user
I Users trust that the information submitted or registeredabout them will be used for filtering purposes, however theirinformation can be used for purposes different than filteringconfiguring an exposure risk. [LFR06]
log files
Recommendation component
Interaction log component
feedback
item suggestionsTraining component
Prediction component
Recommendation server
![Page 8: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/8.jpg)
Privacy considerations
I Recommender systems gather information about users andstore it in a centralized entity, then they apply heuristics ordata mining techniques to learn the users’ interests with thepurpose of detecting which elements are relevant for the user
I Users trust that the information submitted or registeredabout them will be used for filtering purposes, however theirinformation can be used for purposes different than filteringconfiguring an exposure risk. [LFR06]
log files
Recommendation component
Interaction log component
feedback
item suggestionsTraining component
Prediction component
Recommendation server
![Page 9: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/9.jpg)
Privacy considerations
According to [Fon99], keeping user profile information on acentralized entity can lead to exposure risks configured in fiveways:
I Deception by the recipient: The system can lie about itsprivacy policies.
I Mission creep: The system expands its goals in a previouslyunforeseen manner, changing the use of personal informationfor other purposes related to the new goals of theorganization.
I Accidental disclosure: Information about users can be madeavailable accidentally.
I Disclosure by malicious intent: Storage security breachedstealing personal information.
I Forced disclosure: Systems must disclose the information forlegal reasons.
![Page 10: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/10.jpg)
Proposed architecture
Client-side recommender systems is a privacy-per-architecturesolution to avoid exposure scenarios:
I Keep user profile information in user’s device
I Don’t reveal user ratings
![Page 11: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/11.jpg)
Proposed architecture
Client-side recommender systems is a privacy-per-architecturesolution to avoid exposure scenarios:
I Keep user profile information in user’s device
I Don’t reveal user ratings
![Page 12: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/12.jpg)
Outline
Motivation: Privacy in recommender systemsRecommender SystemsPrivacy considerations
A client-side agent for recommendationAppliying client-side predictive modelsContent-based modelCollaborative Filtering model (CF)Hybrid prediction under expert advice
Final considerations
![Page 13: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/13.jpg)
Online Client-based predictive models
I Rating prediction task on client, tested with Movielens10Mdataset, possible ratings restricted to O = {1, 2, 3, 4, 5}.
I Hybrid modelI Content-Based model (CB)I Collaborative Filtering model(CF)
![Page 14: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/14.jpg)
Outline
Motivation: Privacy in recommender systemsRecommender SystemsPrivacy considerations
A client-side agent for recommendationAppliying client-side predictive modelsContent-based modelCollaborative Filtering model (CF)Hybrid prediction under expert advice
Final considerations
![Page 15: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/15.jpg)
Content-based model profiles and prediction
I Content-based filtering (CB): Items are described byfeatures or characteristics of the items to find out therelevance for the user.
Star wars
actor:harrison_ford actor:james_earl_jones actor:mark_hamill actor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucas genre:Adventuregenre:Actiongenre:Sci-Fi
actor:kathy_griffinactor:uma_thurmanactor:bruce_willisactor:christopher_walkenactor:samel_l_jacksonactor:john_travoltagenre:Crimegenre:Comedy
I User has a list of frequent concepts (keywords) Cu , items aredescribed as well by keywords Ci .
I Each user has |O| vectors wou ∈ R|Cu | (o ∈ O)
I mui (Ci × Cu) → R|Cu | binary vector (mui [f ] = 1Cu [f ]∈Ci)
I Rating prediction is: rui =∑
o∈O σ(〈wo ,mui 〉)×o∑o∈O σ(〈wo ,mui 〉)
![Page 16: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/16.jpg)
Content-based model profiles and prediction
I Content-based filtering (CB): Items are described byfeatures or characteristics of the items to find out therelevance for the user.
Star wars
actor:harrison_ford actor:james_earl_jones actor:mark_hamill actor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucas genre:Adventuregenre:Actiongenre:Sci-Fi
actor:kathy_griffinactor:uma_thurmanactor:bruce_willisactor:christopher_walkenactor:samel_l_jacksonactor:john_travoltagenre:Crimegenre:Comedy
I User has a list of frequent concepts (keywords) Cu , items aredescribed as well by keywords Ci .
I Each user has |O| vectors wou ∈ R|Cu | (o ∈ O)
I mui (Ci × Cu) → R|Cu | binary vector (mui [f ] = 1Cu [f ]∈Ci)
I Rating prediction is: rui =∑
o∈O σ(〈wo ,mui 〉)×o∑o∈O σ(〈wo ,mui 〉)
![Page 17: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/17.jpg)
Content-based model profiles and prediction
I Content-based filtering (CB): Items are described byfeatures or characteristics of the items to find out therelevance for the user.
Star wars
actor:harrison_ford actor:james_earl_jones actor:mark_hamill actor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucas genre:Adventuregenre:Actiongenre:Sci-Fi
actor:kathy_griffinactor:uma_thurmanactor:bruce_willisactor:christopher_walkenactor:samel_l_jacksonactor:john_travoltagenre:Crimegenre:Comedy
I User has a list of frequent concepts (keywords) Cu , items aredescribed as well by keywords Ci .
I Each user has |O| vectors wou ∈ R|Cu | (o ∈ O)
I mui (Ci × Cu) → R|Cu | binary vector (mui [f ] = 1Cu [f ]∈Ci)
I Rating prediction is: rui =∑
o∈O σ(〈wo ,mui 〉)×o∑o∈O σ(〈wo ,mui 〉)
![Page 18: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/18.jpg)
Content-based model profiles and prediction
I Content-based filtering (CB): Items are described byfeatures or characteristics of the items to find out therelevance for the user.
Star wars
actor:harrison_ford actor:james_earl_jones actor:mark_hamill actor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucas genre:Adventuregenre:Actiongenre:Sci-Fi
actor:kathy_griffinactor:uma_thurmanactor:bruce_willisactor:christopher_walkenactor:samel_l_jacksonactor:john_travoltagenre:Crimegenre:Comedy
I User has a list of frequent concepts (keywords) Cu , items aredescribed as well by keywords Ci .
I Each user has |O| vectors wou ∈ R|Cu | (o ∈ O)
I mui (Ci × Cu) → R|Cu | binary vector (mui [f ] = 1Cu [f ]∈Ci)
I Rating prediction is: rui =∑
o∈O σ(〈wo ,mui 〉)×o∑o∈O σ(〈wo ,mui 〉)
![Page 19: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/19.jpg)
Content-based model training
I How Ci and Cu are calculated?
I Ci expert knowledge (IMDB.com, rottentomatoes.com)[CBK11]
I Cu concepts the user has interacted at least N times based ona min-count sketch structure [DSHK08][MHS+13]
I How wou is updated?
I Online logistic regression on each vectorI Decreasing learning rate: γt = γ0(1 + αγ0t)−c
I update: wou ← wo
u − γ(tu)(σ(〈wo ,mui 〉)− 1rui=o)mui
![Page 20: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/20.jpg)
Content-based model training
I How Ci and Cu are calculated?I Ci expert knowledge (IMDB.com, rottentomatoes.com)
[CBK11]I Cu concepts the user has interacted at least N times based on
a min-count sketch structure [DSHK08][MHS+13]
I How wou is updated?
I Online logistic regression on each vectorI Decreasing learning rate: γt = γ0(1 + αγ0t)−c
I update: wou ← wo
u − γ(tu)(σ(〈wo ,mui 〉)− 1rui=o)mui
![Page 21: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/21.jpg)
Content-based model training
I How Ci and Cu are calculated?I Ci expert knowledge (IMDB.com, rottentomatoes.com)
[CBK11]I Cu concepts the user has interacted at least N times based on
a min-count sketch structure [DSHK08][MHS+13]
I How wou is updated?
I Online logistic regression on each vectorI Decreasing learning rate: γt = γ0(1 + αγ0t)−c
I update: wou ← wo
u − γ(tu)(σ(〈wo ,mui 〉)− 1rui=o)mui
![Page 22: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/22.jpg)
Content-based model training
I How Ci and Cu are calculated?I Ci expert knowledge (IMDB.com, rottentomatoes.com)
[CBK11]I Cu concepts the user has interacted at least N times based on
a min-count sketch structure [DSHK08][MHS+13]
I How wou is updated?
I Online logistic regression on each vectorI Decreasing learning rate: γt = γ0(1 + αγ0t)−c
I update: wou ← wo
u − γ(tu)(σ(〈wo ,mui 〉)− 1rui=o)mui
![Page 23: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/23.jpg)
Content-based results
I RMSE on dataset, results with N=5, α = 10E−6.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90.95
1
1.05
1.1
1.15
1.2
1.25
1.3
γ0
RM
SE
Metadata predictor RMSE
RMSE trainRMSE cv
![Page 24: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/24.jpg)
Outline
Motivation: Privacy in recommender systemsRecommender SystemsPrivacy considerations
A client-side agent for recommendationAppliying client-side predictive modelsContent-based modelCollaborative Filtering model (CF)Hybrid prediction under expert advice
Final considerations
![Page 25: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/25.jpg)
CF model profiles and predictionI Collaborative Filtering model (CF): Users and items are
described by latent-features, trained from user-iteminteractions
Star wars
q1u
f1 fk
q2u
q3u
q4u
q5u
f1 fk
pi
I Each user has |O| vectors qou ∈ RF (o ∈ O)I An item is described by a vector pi ∈ RF
I Model predicts probability that user u will give rating o toitem i
I Restriction on user profile: qu,f ≥ 0 and∑
o∈O qou,f = 1.I Restriction on item profile:pi,f ≥ 0 and
∑f∈F pi,f = 1
I Probability is πoui = 〈qou , pi 〉I Rating prediction is: rui =
∑o∈O π
oui × o
![Page 26: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/26.jpg)
CF model profiles and predictionI Collaborative Filtering model (CF): Users and items are
described by latent-features, trained from user-iteminteractions
Star wars
q1u
f1 fk
q2u
q3u
q4u
q5u
f1 fk
pi
I Each user has |O| vectors qou ∈ RF (o ∈ O)I An item is described by a vector pi ∈ RF
I Model predicts probability that user u will give rating o toitem i
I Restriction on user profile: qu,f ≥ 0 and∑
o∈O qou,f = 1.I Restriction on item profile:pi,f ≥ 0 and
∑f∈F pi,f = 1
I Probability is πoui = 〈qou , pi 〉I Rating prediction is: rui =
∑o∈O π
oui × o
![Page 27: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/27.jpg)
CF model profiles and predictionI Collaborative Filtering model (CF): Users and items are
described by latent-features, trained from user-iteminteractions
Star wars
q1u
f1 fk
q2u
q3u
q4u
q5u
f1 fk
pi
I Each user has |O| vectors qou ∈ RF (o ∈ O)I An item is described by a vector pi ∈ RF
I Model predicts probability that user u will give rating o toitem i
I Restriction on user profile: qu,f ≥ 0 and∑
o∈O qou,f = 1.
I Restriction on item profile:pi,f ≥ 0 and∑
f∈F pi,f = 1
I Probability is πoui = 〈qou , pi 〉I Rating prediction is: rui =
∑o∈O π
oui × o
![Page 28: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/28.jpg)
CF model profiles and predictionI Collaborative Filtering model (CF): Users and items are
described by latent-features, trained from user-iteminteractions
Star wars
q1u
f1 fk
q2u
q3u
q4u
q5u
f1 fk
pi
I Each user has |O| vectors qou ∈ RF (o ∈ O)I An item is described by a vector pi ∈ RF
I Model predicts probability that user u will give rating o toitem i
I Restriction on user profile: qu,f ≥ 0 and∑
o∈O qou,f = 1.I Restriction on item profile:pi,f ≥ 0 and
∑f∈F pi,f = 1
I Probability is πoui = 〈qou , pi 〉I Rating prediction is: rui =
∑o∈O π
oui × o
![Page 29: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/29.jpg)
CF model profiles and predictionI Collaborative Filtering model (CF): Users and items are
described by latent-features, trained from user-iteminteractions
Star wars
q1u
f1 fk
q2u
q3u
q4u
q5u
f1 fk
pi
I Each user has |O| vectors qou ∈ RF (o ∈ O)I An item is described by a vector pi ∈ RF
I Model predicts probability that user u will give rating o toitem i
I Restriction on user profile: qu,f ≥ 0 and∑
o∈O qou,f = 1.I Restriction on item profile:pi,f ≥ 0 and
∑f∈F pi,f = 1
I Probability is πoui = 〈qou , pi 〉
I Rating prediction is: rui =∑
o∈O πoui × o
![Page 30: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/30.jpg)
CF model profiles and predictionI Collaborative Filtering model (CF): Users and items are
described by latent-features, trained from user-iteminteractions
Star wars
q1u
f1 fk
q2u
q3u
q4u
q5u
f1 fk
pi
I Each user has |O| vectors qou ∈ RF (o ∈ O)I An item is described by a vector pi ∈ RF
I Model predicts probability that user u will give rating o toitem i
I Restriction on user profile: qu,f ≥ 0 and∑
o∈O qou,f = 1.I Restriction on item profile:pi,f ≥ 0 and
∑f∈F pi,f = 1
I Probability is πoui = 〈qou , pi 〉I Rating prediction is: rui =
∑o∈O π
oui × o
![Page 31: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/31.jpg)
CF model training
I How qou is updated?
I Stochastic projected regression on each vector [IICM11]I Decreasing learning rate: γt = γ0(1 + αγ0t)−c
I update: qou ← qou + γ(tu)(1rui=o − (〈pi , qou 〉))piqu ←
∏Duser
(qu)
I How pi is updated?I update: pi ← pi + γ(ti )(1− (〈pi , qou 〉))qou
pi ←∏
Ditem(pi )
I Server doesn’t need rui value in order to update pi
![Page 32: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/32.jpg)
CF model training
I How qou is updated?I Stochastic projected regression on each vector [IICM11]I Decreasing learning rate: γt = γ0(1 + αγ0t)−c
I update: qou ← qou + γ(tu)(1rui=o − (〈pi , qou 〉))piqu ←
∏Duser
(qu)
I How pi is updated?
I update: pi ← pi + γ(ti )(1− (〈pi , qou 〉))qoupi ←
∏Ditem
(pi )I Server doesn’t need rui value in order to update pi
![Page 33: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/33.jpg)
CF model training
I How qou is updated?I Stochastic projected regression on each vector [IICM11]I Decreasing learning rate: γt = γ0(1 + αγ0t)−c
I update: qou ← qou + γ(tu)(1rui=o − (〈pi , qou 〉))piqu ←
∏Duser
(qu)
I How pi is updated?I update: pi ← pi + γ(ti )(1− (〈pi , qou 〉))qou
pi ←∏
Ditem(pi )
I Server doesn’t need rui value in order to update pi
![Page 34: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/34.jpg)
Collaborative Filtering results
I RMSE on dataset, α = 10E−6 for increasing γ0.
0 5 10 20 30 40 50 60 70 80 90 1001.1
1.15
1.2
1.25
1.3
1.35
1.4
1.45
1.5
1.55
1.6
RMSE evolution across γ0 and F for CF model
F dimension
RM
SE
γ0=0.05 cv
γ0=0.1 cv
γ0=0.25 cv
γ0=0.5 cv
![Page 35: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/35.jpg)
Outline
Motivation: Privacy in recommender systemsRecommender SystemsPrivacy considerations
A client-side agent for recommendationAppliying client-side predictive modelsContent-based modelCollaborative Filtering model (CF)Hybrid prediction under expert advice
Final considerations
![Page 36: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/36.jpg)
Hybrid prediction
I How to use both models for rating prediction ? [BL06]
Predictionqcomponent
Recommendationqcomponent
Client-sideqagent
Starqwars
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
Starqwars
q1uq
f1 fkq
q2uq
q3uq
q4uq
q5uq
piq
CB CF
pit
rui^ rui
^1 2
I `(R×O)→ R loss function that scores a prediction
I Cumulative regret: RE ,n =n∑
t=1
(`(pi ,t , ri ,t)− `(rEi ,t , ri ,t)
)I Expert weight: WE ,t−1 =
exp(ηtRE ,t−1)∑e∈E exp(ηtRe,t−1)
I Final prediction: Weighted average of experts
pi ,t =∑
E∈EWE ,t−1 rEi,t∑
E∈EWE ,t−1
![Page 37: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/37.jpg)
Hybrid prediction
I How to use both models for rating prediction ? [BL06]
Predictionqcomponent
Recommendationqcomponent
Client-sideqagent
Starqwars
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
Starqwars
q1uq
f1 fkq
q2uq
q3uq
q4uq
q5uq
piq
CB CF
pit
rui^ rui
^1 2
I `(R×O)→ R loss function that scores a prediction
I Cumulative regret: RE ,n =n∑
t=1
(`(pi ,t , ri ,t)− `(rEi ,t , ri ,t)
)I Expert weight: WE ,t−1 =
exp(ηtRE ,t−1)∑e∈E exp(ηtRe,t−1)
I Final prediction: Weighted average of experts
pi ,t =∑
E∈EWE ,t−1 rEi,t∑
E∈EWE ,t−1
![Page 38: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/38.jpg)
Hybrid prediction
I How to use both models for rating prediction ? [BL06]
Predictionqcomponent
Recommendationqcomponent
Client-sideqagent
Starqwars
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
Starqwars
q1uq
f1 fkq
q2uq
q3uq
q4uq
q5uq
piq
CB CF
pit
rui^ rui
^1 2
I `(R×O)→ R loss function that scores a prediction
I Cumulative regret: RE ,n =n∑
t=1
(`(pi ,t , ri ,t)− `(rEi ,t , ri ,t)
)
I Expert weight: WE ,t−1 =exp(ηtRE ,t−1)∑e∈E exp(ηtRe,t−1)
I Final prediction: Weighted average of experts
pi ,t =∑
E∈EWE ,t−1 rEi,t∑
E∈EWE ,t−1
![Page 39: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/39.jpg)
Hybrid prediction
I How to use both models for rating prediction ? [BL06]
Predictionqcomponent
Recommendationqcomponent
Client-sideqagent
Starqwars
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
Starqwars
q1uq
f1 fkq
q2uq
q3uq
q4uq
q5uq
piq
CB CF
pit
rui^ rui
^1 2
I `(R×O)→ R loss function that scores a prediction
I Cumulative regret: RE ,n =n∑
t=1
(`(pi ,t , ri ,t)− `(rEi ,t , ri ,t)
)I Expert weight: WE ,t−1 =
exp(ηtRE ,t−1)∑e∈E exp(ηtRe,t−1)
I Final prediction: Weighted average of experts
pi ,t =∑
E∈EWE ,t−1 rEi,t∑
E∈EWE ,t−1
![Page 40: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/40.jpg)
Hybrid prediction
I How to use both models for rating prediction ? [BL06]
Predictionqcomponent
Recommendationqcomponent
Client-sideqagent
Starqwars
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
actor:harrison_fordqactor:james_earl_jonesqactor:mark_hamillqactor:alec_guinnessactor:denis_lawsonactor:carrie_fisherdirector:george_lucasqgenre:Adventuregenre:Actiongenre:Sci-Fi
Starqwars
q1uq
f1 fkq
q2uq
q3uq
q4uq
q5uq
piq
CB CF
pit
rui^ rui
^1 2
I `(R×O)→ R loss function that scores a prediction
I Cumulative regret: RE ,n =n∑
t=1
(`(pi ,t , ri ,t)− `(rEi ,t , ri ,t)
)I Expert weight: WE ,t−1 =
exp(ηtRE ,t−1)∑e∈E exp(ηtRe,t−1)
I Final prediction: Weighted average of experts
pi ,t =∑
E∈EWE ,t−1 rEi,t∑
E∈EWE ,t−1
![Page 41: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/41.jpg)
Exponential weighted regret results
I RMSE on dataset, α = 10E−6 for increasing γ0. γ0 of CBhybrid model set to 0.75.
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.91.05
1.1
1.15
1.2
1.25
1.3
1.35
1.4
1.45
1.5RMSE on test set
γ0
RM
SE
RMSE CB modelRMSE CF modelRMSE Hybrid model
![Page 42: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/42.jpg)
Summary
I Client-side agents help the avoidance of user exposure risks .
I Placed in an online learning setting, hybridization of clientside predictive models helps to increase the predictiveperformance of the single models.
I OutlookI Actual model still reveals implicit interaction to
recommendation server.
![Page 43: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/43.jpg)
Client-side hybrid rating prediction forrecommendation
Andres Moreno12 Harold Castro 1 Michel Riveill 2
1School of EngineeringUniversidad de los Andes, Bogota, Colombia
2I3SUniversite de Nice Sophia Antipolis, France
UMAP, 2014
![Page 44: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/44.jpg)
References I
[BL06] Nicolo C. Bianchi and Gabor Lugosi, Prediction, learning, andgames, Cambridge University Press, New York, NY, USA,2006.
[CBK11] Ivan Cantador, Peter Brusilovsky, and Tsvi Kuflik, 2ndworkshop on information heterogeneity and fusion inrecommender systems (hetrec 2011), Proceedings of the 5thACM conference on Recommender systems (New York, NY,USA), RecSys 2011, ACM, 2011.
[DSHK08] Xenofontas Dimitropoulos, Marc Stoecklin, Paul Hurley, andAndreas Kind, The eternal sunshine of the sketch datastructure, Comput. Netw. 52 (2008), no. 17, 3248–3257.
[Fon99] Leonard N. Foner, Political artifacts and personal privacy:The yenta Multi-Agent distributed matchmaking system,Ph.D. thesis, Program in Media Arts and Sciences, School ofArchitecture and Planning, Massachusetts Institute ofTechnology, June 1999.
![Page 45: Client-side hybrid rating prediction for recommendation · Client-side hybrid rating prediction for recommendation Andr es Moreno12 Harold Castro 1 Michel Riveill 2 1School of Engineering](https://reader033.vdocument.in/reader033/viewer/2022042400/5f0e99087e708231d44001d5/html5/thumbnails/45.jpg)
References II[IICM11] Sibren Isaacman, Stratis Ioannidis, Augustin Chaintreau, and
Margaret Martonosi, Distributed rating prediction in usergenerated content streams, Proceedings of the fifth ACMconference on Recommender systems (New York, NY, USA),RecSys ’11, ACM, 2011, pp. 69–76.
[LFR06] Shyong Lam, Dan Frankowski, and John Riedl, Do you trustyour recommendations? an exploration of security and privacyissues in recommender systems, Emerging Trends inInformation and Communication Security (Gunter Muller,ed.), Lecture Notes in Computer Science, vol. 3995, SpringerBerlin / Heidelberg, Berlin, Heidelberg, 2006, pp. 14–29.
[MHS+13] H. Brendan McMahan, Gary Holt, D. Sculley, Michael Young,Dietmar Ebner, Julian Grady, Lan Nie, Todd Phillips, EugeneDavydov, Daniel Golovin, Sharat Chikkerur, Dan Liu, MartinWattenberg, Arnar M. Hrafnkelsson, Tom Boulos, and JeremyKubica, Ad click prediction: A view from the trenches,Proceedings of the 19th ACM SIGKDD InternationalConference on Knowledge Discovery and Data Mining (NewYork, NY, USA), KDD ’13, ACM, 2013, pp. 1222–1230.