SPAR ARQL: QL: More e than n a Shotgun tgun Weddin ing? 8 th - - PowerPoint PPT Presentation

▶
spar arql ql more e than n a shotgun tgun weddin ing
SMART_READER_LITE
LIVE PREVIEW

SPAR ARQL: QL: More e than n a Shotgun tgun Weddin ing? 8 th - - PowerPoint PPT Presentation

Jun 26, 2023 224 likes •513 views

June 5 2014 Recommender ommender Systems ms and SPAR ARQL: QL: More e than n a Shotgun tgun Weddin ing? 8 th Alberto Mendelzon International Workshop Cartagena, Colombia (AMW 2014) Victor Anthony Arras ascue Ayala Martin Przyjac

slide-1
SLIDE 1

Recommender

  • mmender Systems

ms and SPAR ARQL: QL: More e than n a Shotgun tgun Weddin ing?

8th Alberto Mendelzon International Workshop Cartagena, Colombia (AMW 2014)

Victor Anthony Arras ascue Ayala Martin Przyjac aciel-Zab ablocki Thomas as Hornu nung ng Alexan ander Schätzle Georg Lausen University of Freiburg Databases & Information Systems

June 5 2014

slide-2
SLIDE 2
  • 1. Motivation
  • 2. RecSPARQL
  • 3. Experiments
  • 4. Summary

Recommender Systems and SPARQL: More than a Shotgun Wedding?

Overview erview

2

  • 1. Motivation
slide-3
SLIDE 3

 RDF’s flexibility

  • Recommendation domain

 Users, Items, Ratings  Consumption relationship

Recommender Systems and SPARQL: More than a Shotgun Wedding? 3

  • 1. Motivation

Motivat ivatio ion

slide-4
SLIDE 4

 How to get recommendations from RDF-graphs:

SPARQL?

  • Retrieval of explicit data from RFD-graphs
  • Graph pattern matching
  • Flexible
  • No possible to express fuzzy queries

 Similarity

Recommender Systems and SPARQL: More than a Shotgun Wedding? 4

  • 1. Motivation

Motivat ivatio ion

SPARQL

Source: http://courses.ischool.berkeley.edu/i253/f11/

slide-5
SLIDE 5

Recommender Systems and SPARQL: More than a Shotgun Wedding? 5

  • 1. Motivation

Motivat ivatio ion

Recommender Systems

 How to get recommendations from RDF-graphs:

Recommender Systems

  • Predicts a degree of preference for a user towards a set of

non-consumed items

  • Based on Information Retrieval techniques

Source: http://courses.ischool.berkeley.edu/i253/f11/

slide-6
SLIDE 6

Recommender Systems and SPARQL: More than a Shotgun Wedding? 6

Motivat ivatio ion

 A straightforward approach

  • 1. Motivation

 Classic recommender

  • Inpu

put: users, items, ratings

  • Outpu

put: users, recommended items, predicted rating Recommendations

slide-7
SLIDE 7

7

Motivat ivatio ion

 Example: lack of flexibility

  • Collaborative filtering approach

 Recommendations from similar users

  • 1. Motivation

1 3 5 0 2 1 0 4 1 1 3 2 0 0 3 0 3 4 1 5 1 3 5 0 1 2 1 2 3 2 Recommender Systems and SPARQL: More than a Shotgun Wedding?

(A) Extraction / Pre-processing SPARQL (B) Recommendation System (CF)

slide-8
SLIDE 8

Recommender Systems and SPARQL: More than a Shotgun Wedding? 8

Motivat ivatio ion

 Example: lack of flexibility

  • Collaborative filtering approach

 Recommendations from similar users

  • 1. Motivation

U1 U5 U17 U43 U55

1 3 5 0 2 1 0 4 1 1 3 2 0 0 3 0 3 4 1 5 1 3 5 0 1 2 1 2 3 2

(B) Recommendation System (CF)

slide-9
SLIDE 9

Recommender Systems and SPARQL: More than a Shotgun Wedding? 9

Motivat ivatio ion

 Example: lack of flexibility

  • Collaborative approach

 Recommendations from similar users

  • 1. Motivation

U1 U5 U17 U43 U55

  • Customization

 Neighbors geographically close  Age of neighbors differ by 𝜀  Speak same languages  etc….

slide-10
SLIDE 10

Recommender Systems and SPARQL: More than a Shotgun Wedding? 10

Motivat ivatio ion

 Example: lack of flexibility

  • Collaborative approach

 Recommendations from similar users

  • 1. Motivation

U1 U5 U17 U43 U55

  • Customization

 Neighbors geographically close  Age of neighbors differ by 𝜀  Speak same languages  etc….

 Where is the problem?

  • Fixed recommender model
slide-11
SLIDE 11
  • 1. Motivation
  • 2. RecSPARQL
  • 3. Experiments
  • 4. Summary

Recommender Systems and SPARQL: More than a Shotgun Wedding?

  • 2. Appr

proach

Overview erview

11

slide-12
SLIDE 12

Recommender Systems and SPARQL: More than a Shotgun Wedding? 12

Our Ap Approach

  • ach

 RecSesame

  • Platform for the evaluation of

RecSPARQL queries

 Based on Sesame’s framework  Cache System

  • 2. Appr

proach

 RecSPARQL: Recommendations + SPARQL

  • Extension of SPARQL 1.1

 Consistent mechanism to select parts of the graph  Flexibility

slide-13
SLIDE 13

Recommender Systems and SPARQL: More than a Shotgun Wedding? 13

RecSPA cSPARQ RQL in a nutshe shell

  • 2. Appr

proach

RECOMMEND [Projected Variables] USING [Recommendation Algorithm] WHERE { [Basic Graph Pattern] } BASED ON { [RecSPARQL Type Pattern] [RecSPARQL Model Building Pattern] }

slide-14
SLIDE 14

Recommender Systems and SPARQL: More than a Shotgun Wedding? 14

RecSPA cSPARQ RQL in a nutshe shell

  • 2. Appr

proach

RECOMMEND [Projected Variables] USING [Recommendation Algorithm] WHERE { [Basic Graph Pattern] } BASED ON { [RecSPARQL Type Pattern] [RecSPARQL Model Building Pattern] }

  • Must contain

recommendation entities

slide-15
SLIDE 15

Recommender Systems and SPARQL: More than a Shotgun Wedding? 15

RecSPA cSPARQ RQL in a nutshe shell

  • 2. Appr

proach

RECOMMEND [Projected Variables] USING [Recommendation Algorithm] WHERE { [Basic Graph Pattern] } BASED ON { [RecSPARQL Type Pattern] [RecSPARQL Model Building Pattern] }

  • Algorithm used

 Algorihtms

  • Content-based (CB)
  • Collaborative Filtering (CF)
  • Hybrid (H)
slide-16
SLIDE 16

Recommender Systems and SPARQL: More than a Shotgun Wedding? 16

RecSPA cSPARQ RQL in a nutshe shell

  • 2. Appr

proach

RECOMMEND [Projected Variables] USING [Recommendation Algorithm] WHERE { [Basic Graph Pattern] } BASED ON { [RecSPARQL Type Pattern] [RecSPARQL Model Building Pattern] }

  • Similarity

criteria

 Content-based

  • Genre
  • Cast
  • Director

 Collaborative filtering

  • Watched movies and ratings
  • Age
  • Geographical location
slide-17
SLIDE 17

 RecSPARQL in a nutshell

Recommender Systems and SPARQL: More than a Shotgun Wedding? 17

RecSPA cSPARQ RQL in a nutshe shell

  • 2. Appr

proach

RECOMMEND [Projected Variables] USING [Recommendation Algorithm] WHERE { [Basic Graph Pattern] } BASED ON { [RecSPARQL Type Pattern] [RecSPARQL Model Building Pattern] }

  • Similar to

projection

  • Makes it possible

to project recommendations

  • ?movi

vie.REC

slide-18
SLIDE 18

 Filters

  • Recommend only action movies

Recommender Systems and SPARQL: More than a Shotgun Wedding? 18

RecSPARQL’s flexi xibi bility lity

  • 2. Appr

proach

FILTER ( ?genre.REC = “action”)

?genre.REC EC ?genre.REC EC

slide-19
SLIDE 19

 Filters

  • Recommend only from users whose age differ by at most 5

years

Recommender Systems and SPARQL: More than a Shotgun Wedding? 19

RecSPARQL’s flexi xibi bility lity

  • 2. Appr

proach

FILTER ( abs(xsd:integer(?age) - xsd:integer(?age.REC)) <= 5)

?age.REC EC ?age

slide-20
SLIDE 20

 Much more…

  • Recommend movies watched under a certain context
  • Recommend movies whose directors have the same citizenship
  • f the user for which we want the recommendations
  • Recommend ….

Recommender Systems and SPARQL: More than a Shotgun Wedding? 20

RecSPARQL’s flexi xibi bility lity

  • 2. Appr

proach

FILTER ( ?watchTime.REC = “weekend” && ?company.REC = “partner”) .

slide-21
SLIDE 21
  • 1. Motivation
  • 2. RecSPARQL
  • 3. Experiments
  • 4. Summary

Recommender Systems and SPARQL: More than a Shotgun Wedding?

  • 4. Experime

ments

Experi riment ments

21

slide-22
SLIDE 22

Recommender Systems and SPARQL: More than a Shotgun Wedding? 22

Experi riment ments

 Example: lack of flexibility

  • Collaborative approach

 Recommendations from similar users

  • 1. Motivation

U1 U5 U17 U43 U55

 Gradual restriction of

Neighborhood:

  • Neighbors geographically close
  • Age of neighbors differ by 𝜀
slide-23
SLIDE 23

Recommender Systems and SPARQL: More than a Shotgun Wedding? 23

Experi riment ments

 Example: lack of flexibility

  • Collaborative approach

 Recommendations from similar users

  • 1. Motivation

U1 U5 U17 U43 U55

FILTER ( abs(xsd:integer(?age) - xsd:integer(?age.REC)) <= %K%) . FILTER ( abs(xsd:integer(?zip) - xsd:integer(?zip.REC)) <= %L% )

slide-24
SLIDE 24

Recommender Systems and SPARQL: More than a Shotgun Wedding? 24

Experi riment ments

 Benefitial

  • Backed by our experiments
  • 1. Motivation

U1 U5 U17 U43 U55

  • Similarity among users

in the neighborhood

 Increases

  • Customization

 Neighbors geographically close  Age of neighbors differ by 𝜀

slide-25
SLIDE 25

Recommender Systems and SPARQL: More than a Shotgun Wedding? 25

Experi riment ments

 Benefitial

  • Backed by our experiments
  • 1. Motivation

U1 U5 U17 U43 U55

  • Similarity among users

in the neighborhood

 Increases

  • Customization

 Neighbors geographically close  Age of neighbors differ by 𝜀

1 3 5 0 2 1 0 4 1 1 3 2 0 0 3

0 3 4 1 5 1 3 5 0 1 2 1 2 3 2

slide-26
SLIDE 26

 Tight integration of recommender systems with SPARQL  Customizable recommendations on arbitrary RDF graphs  Futu

ture e Work

  • Enhance the integration of both paradigms
  • Support more recommendations techniques
  • Increase the expressiveness of RecSPARQL

 Binding variables  Sub-queries  Property paths

Recommender Systems and SPARQL: More than a Shotgun Wedding? 26

  • 5. Summa

mmary

Summary mary

  • V. A. Arrascue Ayala, M. Przyj

yjaciel-Zablocki, T. Hornung ung, A. Schä hätzle, G. Laus usen, n, Extend ending ng SPARQL QL for Recommend endations

  • ns. In SWIM

M (ACM M SIGMOD MOD), pages 1-8, 2014.

slide-27
SLIDE 27

Recommender Systems and SPARQL: More than a Shotgun Wedding?

Thanks ks

Thanks nks for your r atten ention tion!

27