New sciences for a new web Prabhakar Raghavan Yahoo! Research - - PowerPoint PPT Presentation

new sciences for a new web
SMART_READER_LITE
LIVE PREVIEW

New sciences for a new web Prabhakar Raghavan Yahoo! Research - - PowerPoint PPT Presentation

New sciences for a new web Prabhakar Raghavan Yahoo! Research Microeconomics Social Sciences Statistics Computation 1 What companies like mine do $2 Intent Marketplace Info/service Marketplace $1 Real-time matching Audience arrives


slide-1
SLIDE 1

1

New sciences for a new web

Prabhakar Raghavan Yahoo! Research

Computation

Microeconomics Social Sciences Statistics

slide-2
SLIDE 2

Yahoo! Research

2

What companies like mine do

Audience arrives Info and Service providers – advertisers, publishers, commerce $2

  • $5

$1

Info/service Marketplace

Real-time matching

Intent Marketplace

slide-3
SLIDE 3

Yahoo! Research

3

The promise

  • Measurability – Auditable notion of

audience engagement and fulfillment

  • Targetability – matching good enough to

let marketplace design dictate granularity

  • Scale – leads to virtuous cycle in system

learning, as well thicker markets for monetization

slide-4
SLIDE 4

Yahoo! Research

4

Algorithmic results =Audience Advertisements =Monetization

slide-5
SLIDE 5

Yahoo! Research

5

Search: Content vs. Intent

  • Premise:

–People don’t want to search –People want to get tasks done

I want to book a vacation in Tuscany. Start Finish

Broder 2002, A Taxomony of web search

slide-6
SLIDE 6

Yahoo! Research

6

slide-7
SLIDE 7

Yahoo! Research

7

slide-8
SLIDE 8

Yahoo! Research

8

Web of objects

  • Documents

–Object extraction –Long tail normalization

  • Queries

–Intent extraction –(Matching) object assembly

  • 1. The Savoy

Located on The Strand in the

  • f the West End theatre distric

hotel near leicester square

Go

Reviews … Deals …

slide-9
SLIDE 9

Yahoo! Research

9

Grand challenges

  • What’s an intent?
  • What’s a web of objects?

–Social annotations, geo …

  • How do you gauge intent satisfaction?

–How do you build a framework for relevance?

  • Uniformity of experience vs. Diversity
  • f intent fulfillment
slide-10
SLIDE 10

Yahoo! Research

10

A post-HCI revolution

  • Science of online audience

engagement

–Not just about people interacting with computers –But about people interacting with other people, information and services –An intrinsically data-driven science

slide-11
SLIDE 11

Yahoo! Research

11

Prototypical insights

  • Why do people choose to lurk or participate?
  • Why do people create new online personas?
  • Why are YouTube, Facebook and Flickr

successful? (and many others, not)

  • What new genres are emerging - and what

can we provoke?

– For content creation – Enrichment – Participation

slide-12
SLIDE 12

Yahoo! Research

12

Audience engagement

  • What does it mean to have an

engaged audience?

–Broadcast: hours watched/listened –Print: Circulation

  • Today on the internet: page views,

hours …

  • Who cares?

–Advertisers, publishers

slide-13
SLIDE 13

Yahoo! Research

13

Proxies for engagement

  • Page views, repeat visits
  • Time elapsed within page and

between pages

  • Clicks, click-throughs and click chains
  • Content generation
  • Content transport through networks
  • Subscriptions
  • Creation of user IDs
  • Creation of user profiles
  • Downloads, purchases

Increasing engagement

slide-14
SLIDE 14

Yahoo! Research

14

Audience metrics

  • How about
  • Where did this come from?

–I made it up, not totally implausible –But utterly ungrounded in psychology

) _ log( ) _ (

  • d

neighborho user spent time repeat

pageviews

× ×

α

Granovetter, Schelling 1978; Kempe/Kleinberg/Tardos 2003; Domingos/Richardson 2002

slide-15
SLIDE 15

Yahoo! Research

15

Grand challenges

  • Devise and standardize defensible

metrics of online engagement

  • Use these to predictively design online

experiences

–Not a subsitute for creativity –But a scientific basis

slide-16
SLIDE 16

Yahoo! Research

16

The Long Tail (The “power law distribution” …)

  • The expected number
  • f times a query is

asked is a constant

– But the mass in the tail is non-trivial.

Most of our math builds on small- tailed distributions (binomials, Normal, Poisson ...) Most of our industry encounters long-tailed distributions (power laws ...)

slide-17
SLIDE 17

Yahoo! Research

17

Estimating user responses

  • What is the likely response of a user to

a page view?

–(Response prediction)

  • What if we’ve almost never seen such

a user before

–But can’t afford to ignore him Fresh challenges in building and

  • ptimizing computing artifacts.
slide-18
SLIDE 18

Yahoo! Research

18

Learning in long-tailed spaces

Features Long-tailed sparsity Domain info Hierarchical bandits, Block estimation

slide-19
SLIDE 19

Yahoo! Research

19

Exchange and marketplace design

  • Today providers vie for intents

– Males 25-30 in NYC during the super bowl – Women shopping for $1000+ dishwashers

  • Risk management

–Bid on either spot or future contracts –Charge for impressions, clicks or actions … arbitrage

slide-20
SLIDE 20

Yahoo! Research

20

Design questions

  • How do you express a bid?
  • How do you price a bid?

–Antecedent in airline yield management –But different here for several reasons

  • Both supply and demand are stochastic
  • Near (and not-so-near) substitutes
  • Uniformity/fairness desires
  • How do you match a bid to audience?
slide-21
SLIDE 21

Yahoo! Research

21

A new convergence

  • Monetization and user engagement an

intrinsic part of system design

–Not an afterthought –Mistakes are costly!

  • Computing meets humanities like

never before – sociology, economics, anthropology …

slide-22
SLIDE 22

22

Thank you.

Questions?

pragh@yahoo-inc.com http://research.yahoo.com