ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos...

18
ValuePick: Towards a Value-Oriented Dual- Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia

Transcript of ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos...

Page 1: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

ValuePick: Towards a Value-Oriented Dual-Goal Recommender System

Leman Akoglu Christos Faloutsos

OEDM in conjunction with ICDM 2010 Sydney, Australia

Page 2: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Recommender Systems

Traditional recommender systems try to achieve high user satisfaction

2 of 19

Page 3: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Dual-goal Recommender Systems

Dual-goal recommender systems try to achieve (1) high user satisfaction as well as(2) high-“value” vendor gain

-“value”

Trade-off user

satisfaction vs.

vendor profit

3 of 19

Page 4: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

vertices ranked by proximity

v253v162v261v327

.

.

.

Dual-goal Recommender Systems

network-“value”

4 of 19

query vertex

Page 5: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

v253v162v261v327

.

.

.

Dual-goal Recommender Systems vertices ranked by

proximity

network-“value”

5 of 19

Page 6: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Dual-goal Recommender Systems

v253v162v261v327

.

.

.

network-“value” vertices ranked by

proximity

network-“value”

Trade-off user satisfaction

vs. network

connectivity 6 of 19

Page 7: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Vendor

Main concerns: We cannot make the highest value

recommendations Recommendations should still reflect

users’ likes relatively well

Dual-goal Recommender Systems

7 of 19

User

Page 8: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

User Vendor

Carefully perturb (change the order of) the proximity-ranked list of recommendations

Controlled by a tolerance for each user

ValuePick: Main idea

ζζ

8 of 19

Page 9: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

ValuePick Optimization Framework“valu

e” proximity Total expected

gain (assuming proximity ~ acceptance prob.)

toleranceϵ [0,1]

average proximity score of original top-k

9 of 19

DETA

ILS

Page 10: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

ValuePick ~ 0-1 Knapsackvalue

maximum weight W allowedweight of item

i

We use CPLEX to solve our integer programming optimization problem

10 of 19

DETA

ILS

Page 11: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Pros and Cons of ValuePickCons: In marketing, it is hard to predict the

effect of an intervention in the marketing scheme, i.e., not clear how users will respond to ‘adjustments’

Pros: Tolerance ζ can flexibly (and even

dynamically) control the `level-of-adjustment’

Users rate same item differently at different times, i.e., users have natural variability in their decisions.

11 of 19

Page 12: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Experimental Setup I Two real networks

Netscience – collaboration network DBLP – co-authorship network

Four recommendation schemes:1) No Gain Optimization (ζ = 0)2) ValuePick (ζ = 0.01, ζ = 0.02)3) Max Gain Optimization (ζ = 1)4) Random

“value” is centrality

12 of 19

Page 13: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Experimental Setup II

Given a recommendation scheme At each step

For each node Make a set of recommendations to node using Node links to node ϵ with prob. proximity(,)

Re-compute proximity and centrality scores

Simulation steps:

We use =5 and =30

13 of 19

Page 14: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Comparison of schemes

ValuePick provides a balance between user satisfaction (high E), and vendor gain (small diameter).

EX

PER

IMEN

TS

14 of 19

Page 15: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Recommend by heuristic

Simple perturbation heuristics do not balance user satisfaction and vendor gain properly.

EX

PER

IMEN

TS

15 of 19

Page 16: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Computational complexityEX

PER

IMEN

TS

16 of 19

Making ValuePick recommendations to a given node involves:1 - finding PPR scores

O(#edges)2 - solving ValuePick optimization w/ CPLEX

1/10 sec. to solve among top 1K nodes

Page 17: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

Conclusions Problem formulation: incorporate the

“value” of recommendations into the system Design of ValuePick:

parsimonious single parameter ζ flexible adjust ζ for each user

dynamically general use any “value” metric

Performance study: experiments to show proper trade of user

acceptance in exchange for higher gain CPLEX with fast solutions

17 of 19

Page 18: ValuePick : Towards a Value-Oriented Dual-Goal Recommender System Leman Akoglu Christos Faloutsos OEDM in conjunction with ICDM 2010 Sydney, Australia.

User Vendor

ζ

THANK YOUwww.cs.cmu.edu/~lakoglu

[email protected]

18 of 19