overview of machine learning opportunities in retail @ svml 2015.01.30

25
PROPRIETARY AND CONFIDENTIAL Overview of Machine Learning Opportunities in Retail Sushant Shankar | Chief Data Scientist | 01/30/2015 Silicon Valley Machine Learning 1

Upload: sushant-shankar

Post on 14-Jul-2015

697 views

Category:

Engineering


2 download

TRANSCRIPT

PROPRIETARY AND CONFIDENTIAL

Overview of Machine Learning Opportunities in RetailSushant Shankar | Chief Data Scientist | 01/30/2015Silicon Valley Machine Learning

1

PROPRIETARY AND CONFIDENTIAL

Agenda

1. Product Overview

2. ML Algorithms for Personalization

3. ML Algorithms for Planning

01-30-15 2

PROPRIETARY AND CONFIDENTIAL

Product Overview

01-30-15 3

PROPRIETARY AND CONFIDENTIAL 4

A typical visit to an e-commerce site is not straight-forward and not conducive to rules

01-30-15

t

Google: ‘Converse shoes’

Purchase!

PROPRIETARY AND CONFIDENTIAL

• Segmentation• Campaigns• A/B tests

Current Tools for E-commerce are highly driven by Rules

5

Rules are manually specifying conditional probabilities!

01-30-15

Rules drive:

PROPRIETARY AND CONFIDENTIAL

The Reflektion Platform leverages Machine Learning to learn the ‘optimal policies’

601-30-15

Implement 1 to 1 experiencesacross devices

Measure performance, identify opportunities and generate insights

Drive lifetime value and incremental traffic

PROPRIETARY AND CONFIDENTIAL

ML Algorithms for Personalization

701-30-15

PROPRIETARY AND CONFIDENTIAL

Want to learn the Response of a User interacting with a Context

801-30-15

Response

User

Context

PROPRIETARY AND CONFIDENTIAL

Typical optimization is maximizing the average responses

01-30-15 9

PROPRIETARY AND CONFIDENTIAL

, w

A better approach is to maximize each user’s responses

01-30-15 10

PROPRIETARY AND CONFIDENTIAL 1101-30-15

Ideally, we would have the users draw us this curve. Realistically, we need to infer this curve.

PROPRIETARY AND CONFIDENTIAL

We can infer this curve through supervised and un-supervised models

1201-30-15

User events Context

Get new experience

New (user, context)

Features (slide 13)

Train models (slide 14,15)

...

...(slide 16)

PROPRIETARY AND CONFIDENTIAL

1. Merchandise2. Brand3. Site4. User demographic5. Core Business Goal

Features need to incorporate domain knowledge

1301-30-15

vs.

User Context

(U, C)

Features

Train

Experience

...

...

PROPRIETARY AND CONFIDENTIAL

Prior

Model Selection is itself a multi-level State Space Search

1501-30-15

Internal Model Evaluation (t)

Data

Properties of Data

Best Models ⊂ Models

Optimal Models

User Context

(U, C)

Features

Train

Experience

...

...

Model Evaluation(s)

Model(s)

Experiments

External Model Evaluation (t)

PROPRIETARY AND CONFIDENTIAL

Need to have over-rides that reflect business considerations

1601-30-15

User Context

(U, C)

Features

Train

Experience

...

...

PROPRIETARY AND CONFIDENTIAL

ML Algorithms for Planning

01-30-15 17

PROPRIETARY AND CONFIDENTIAL

How did you drive results? What insights can you provide?

1801-30-15

1. Businesses need to understand how results were driven.a. Can expose the Machine-learned weights in a digestible way.

2. Can surface these insights into tools to allow businesses to make decisions about/through:a. Merchandise

i. Assortment Planningii. Inventory Forecasting

b. Marketingi. Channel Managementii. User Segmentationiii. Campaign Management

PROPRIETARY AND CONFIDENTIAL

Auto-segmentation of users and contexts

1901-30-15

(Users, Context)

1. Take interesting Users, Contexts, (users, contexts)

2. Cluster (un)successful behaviors together to:a. ‘Personas’ of consumers based on

what are driving KPIsb. Best contextsc. Sort out interesting business

opportunitiesd. Anomalies from expected behavior

PROPRIETARY AND CONFIDENTIAL

5

Predictive models can be used to simulate business decisions

2001-30-15

1

2

3

4

f 12(price, user location,...)

f13(price, user location,...)

f 34 ...

f35 ...

...

...

...

3

4

5

1

2

PROPRIETARY AND CONFIDENTIAL

We are a growing company and always looking for great [email protected]

Questions?

21

PROPRIETARY AND CONFIDENTIAL

Backup Slides

22

PROPRIETARY AND CONFIDENTIAL

Marketing funnel in reality is complex

2301-30-15

Source: http://adamhcohen.com/the-new-marketing-funnel/

PROPRIETARY AND CONFIDENTIAL

At any point in the interaction, there is a (User, Context) state

2401-30-15

PROPRIETARY AND CONFIDENTIAL 25

(Users, Context) Metric

01-30-15

Understanding consumer behavior needs to understand user, context, and response attributes