boosting consumer engagement at paypal

34
Boosting Consumer Engagement at PayPal Sujit Mathew, Goh Yew Yap, Chen Yanhui

Upload: sujmat

Post on 14-Apr-2017

94 views

Category:

Presentations & Public Speaking


2 download

TRANSCRIPT

Page 1: Boosting consumer engagement at PayPal

Boosting Consumer Engagement at PayPal

Sujit Mathew, Goh Yew Yap, Chen Yanhui

Page 2: Boosting consumer engagement at PayPal

PayPal

Page 3: Boosting consumer engagement at PayPal

©2015 PayPal Inc. Confidential and proprietary.

Two decades ago, our founders invented payment technology to make buying and selling faster, safer, and easier—and put economic power where it belongs: In the hands of people.

Page 4: Boosting consumer engagement at PayPal

©2015 PayPal Inc. Confidential and proprietary.

Mass Adoption of Mobile Devices

Digitization of Cash

Transformation of Cards

Fragmentation of Payment Types, Technology and

Channels

Rise of Fraud and Cybercrime

Money is

changing

Page 5: Boosting consumer engagement at PayPal

PayPal is leading the transformation

©2015 PayPal Inc. Confidential and proprietary.

AT SCALE*

173 Million Customers**

$235 Billion TPV

$8 Billion Revenue

4 Billion Transactions

WITH MOMENTUM

+19 Million Customers Gained in 2014

+26% y/y TPV Growth

+22% y/y Transaction Growth

*Stats are full year 2014, unless otherwise noted. **Stat is Q3 2015

Page 6: Boosting consumer engagement at PayPal

©2015 PayPal Inc. Confidential and proprietary.

PayPal processed $46 billion in mobile payment volume in 2014,

up 68% over 2013.

In 2014, 20% of PayPal’s net Total Payment Volume was

from mobile payments.

Venmo processed $2.4B in Total Payment Volume in 2014.

In Q3 2015 Venmo’s Payment Volume was $2.11 billion – up 201% year over year.

In 2014, PayPal and Venmo combined handled billions in P2P payment volume globally and nearly half of that volume was international.

A leader in

person-person payments

Page 7: Boosting consumer engagement at PayPal

Use Case

Page 8: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary. 8

“Boost consumer engagement by recommending merchants and products to

consumers.”

Page 9: Boosting consumer engagement at PayPal

©2015  PayPal  Inc.  Confiden4al  and  proprietary.  

66  Million  individual  payments  processed  for  chari4es  via  PayPal.  

36  Million  consumers  used  PayPal  to  make  a  payment  to  a  charity.  

13%  of  dona4ons  through  PayPal  in  2014  were  made  on  a  mobile  device.  

418,000  chari4es  used  PayPal  to  accept  dona4ons.  

$5.7  Billion  processed  for  chari4es  by  PayPal.  

65%  YoY  growth  in  PayPal’s  total  mobile  payment  value  to  chari4es  globally.  

Bridging Consumers and Charities

 *Stats  are  PayPal  full  year  2014  reports  

Page 10: Boosting consumer engagement at PayPal

Modeling

Page 11: Boosting consumer engagement at PayPal

Overview

Stack

Graph

Collaborative Filtering

Content Model

Deployment

Page 12: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Technology Stack

12

Hadoop / MapReduce

Mahout Pig

Python / Shell

HDFS Cassandra

Titan

Gremlin

Page 13: Boosting consumer engagement at PayPal

Graph Modeling

Page 14: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Build Property Graph Based on P2P transaction data.

14

Discover Communities within P2P Data

Page 15: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Discover Key Influencers

15

Eigenvector Centrality

Page 16: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Property Graphs

G = (V , E , λ)

V = vertices E = Edges λ = Properties

16

Page 17: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Property Graphs

17

User Type Age Influence Score

Charity Type Name

g.V[[type:"Charity"]] .inE("Donate") .filter{it.getProperty('Amount') > 25} .outV.filter{it.getProperty('Influence') > 0.5}

Amount

Recommend

Enrich the graph with Donations and Social data.

Key Interests within group

Charity

Send

Find all Key influencers who have donated more than 25 USD to the charity

Page 18: Boosting consumer engagement at PayPal

Collaborative Filtering

Page 19: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Collaborative Filtering 101

19

Commerce Interaction Matrix

We want to model the affinity between consumers and merchants

More transactions occurred, more confident we believe the relationship

Consumer

nonprofits

Likeness Matrix

Confidence Matrix

Page 20: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Collaborative Filtering 101 A matrix factorization method

20

Data Fitting Regularization

Merchant

nonprofits

C

onsu

mer

C

onsu

mer

d

d

Page 21: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Alternative Least Square

21

Iteratively update Fix V and update U: Fix U and update V:

Regularization Data Fitting

Page 22: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Scalable Collaborative Filtering Improve the scalability of ALS

22

Shard 1 Shard 2 Shard 3 Consumer

Merchant

Commerce Interaction Matrix

Mapreduce Job for Shard 1

Mapreduce Job for Shard 2

Mapreduce Job for Shard 3

Stage 1: Compute the individual contributions of each rating

Stage 2: Aggregate all contributions for every user and update their models in parallel

Reference: http://www.slideshare.net/jekky_yiqun/scalable-collaborative-filtering-for-commerce-recommendation

Global MapReduce Job for ALS

Page 23: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

What does CF Learn?

23

⎥⎥⎥⎥

⎢⎢⎢⎢

⎡ −

⎥⎥⎥⎥

⎢⎢⎢⎢

⎡ −

=⋅

!!!!

""""

03.05.015.09.02.11.03.001.001.03.12.07.0

VUT

Each vector of U model a consumer by d implicit attributes

Each vector of V model a merchant by d implicit attributes

Score = UiT . Vj = 1.162

Page 24: Boosting consumer engagement at PayPal

Content Model

Page 25: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Data & Feature Dictionary cust_id string age range [10:20:30:40:50:] gender categorical [M,F,U] country categorical [NP, IN, CN] spend numeric nonprofits label

Cust ID Age … Gender Country Spend nonprofits

1 28 M NP 10.5 1

2 35 F CN 100 2

3 30 M IN 25.1 1

4 34 F IN 15 3

5 32 M IN 5 4

6 25 F IN 22.5 1

3 30 M IN 12 3

3 30 M IN 1 2

Feature Dictionary Dataset

Name

Type

Values

Page 26: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Training the model

Data Source

Featurizer

Business Logic

Feature Dictionary

Other resources

Features Learner Predictive Models

ML Algorithm

Page 27: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Logistic Regression

Page 28: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Recurrent Neural Network - LSTM

Source: http://colah.github.io/posts/2015-08-Understanding-LSTMs/

Page 29: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Combining the result

Teradata & HDFS

Interaction Matrix

Features

CF Model

Content Model

Hybrid Result

CF Engine

Ensemble

Page 30: Boosting consumer engagement at PayPal

Deployment

Page 31: Boosting consumer engagement at PayPal

•  Emails, Web, Mobile Applications

•  REST APIs Platform

• ML Models Data Science

Page 32: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

REST API

The recommendations will

be stored.

MODEL

The recommendations will be generated

and uploaded.

FRONT-END WEB will query for

recommendations

Recommendations of respective

customer will be returned.

Web Emails Mobile Offers

Page 33: Boosting consumer engagement at PayPal

© 2015 PayPal Inc. All rights reserved. Confidential and proprietary.

Wrap Up

•  Approaches to modeling –  Graph Model –  CF Model –  Content Model

•  Deployment of models

Page 34: Boosting consumer engagement at PayPal

Thank You

Sujit Mathew, Goh Yew Yap, Chen Yanhui