Download - 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research
![Page 1: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/1.jpg)
1
Web SearchFrom Information Retrieval to Microeconomic Modeling
Prabhakar Raghavan
Yahoo! Research
![Page 2: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/2.jpg)
2Yahoo! Research
What is web search?
• Access to “heterogeneous”, distributed information– Heterogeneous in creation– Heterogeneous in accuracy– Heterogeneous in motives
• Multi-billion dollar business– Source of new opportunities in marketing
• Strains the boundaries of trademark and intellectual property laws
• A source of unending technical challenges
![Page 3: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/3.jpg)
3Yahoo! Research
The coarse-level dynamics
Content creators Content aggregators
Feeds
Crawls
Content consumers
Adv
erti
sem
ent
Edi
tori
al
Sub
scri
ptio
nT
rans
acti
on
![Page 4: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/4.jpg)
4Yahoo! Research
![Page 5: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/5.jpg)
5Yahoo! Research
Brief (non-technical) history
• Early keyword-based engines
– Altavista, Excite, Infoseek, Inktomi, Lycos, ca. 1995-1997
• Paid placement ranking: Goto (morphed into Overture Yahoo!)
– Your search ranking depended on how much you paid
– Auction for keywords: casino was expensive!
![Page 6: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/6.jpg)
6Yahoo! Research
Brief (non-technical) history
• 1998+: Link-based ranking pioneered by Google
– Blew away all early engines except Inktomi
– Great user experience in search of a business model
– Meanwhile Goto/Overture’s annual revenues were nearing $1 billion
![Page 7: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/7.jpg)
7Yahoo! Research
Brief (non-technical) history
• Result: Google added “paid-placement” ads to the side, separate from search results
• 2003: Yahoo follows suit, acquiring Overture (for paid placement) and Inktomi (for search)
![Page 8: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/8.jpg)
8Yahoo! Research
Algorithmic results CPC Advertisements
![Page 9: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/9.jpg)
9Yahoo! Research
Editorial
User reviews
Ads
![Page 10: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/10.jpg)
10Yahoo! Research
Types of content
• Editorial content: books, music, professionally-produced websites
• User-generated content: blogs, reviews, bulletin boards, groups, etc.
• Total web growth: 1-3M pages/day– Not “real” growth– Think text content…
Courtesy: Andrew Tomkins
![Page 11: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/11.jpg)
11Yahoo! Research
Total text content
• 6B people type 4 hrs/day at 100 wpm
• Storage: 52PB/yr = Cost: $25M/yr
• In another 5 years, this looks about like the cost of having 10 people on your payroll
• Conclusion 1: any company with tens of people can store every bit of text produced by every human on the planet
• Conclusion 2: no scale-based differentiation around text content
(of course, not all content is text…)Courtesy: Andrew Tomkins
![Page 12: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/12.jpg)
12Yahoo! Research
User generated content (UGC)
• New content– 2 billion pages of editorial content– Tiny number of songs, etc– 5-10 billion pages UGC exist (already ~10%
of consumption), growing
• Note: UGC did not exist as a category a couple of years ago!– Rapidly becoming a key growth area of
consumed web content – but we don’t know how to process it!
Courtesy: Andrew Tomkins
![Page 13: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/13.jpg)
13
Tags The simplest form of UGC
Is the Turing test always the right question?
![Page 14: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/14.jpg)
14Yahoo! Research
![Page 15: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/15.jpg)
15Yahoo! Research
The power of social tagging
• Flickr – community phenomenon
• Millions of users share and tag each others’ photographs (why???)
• The wisdom of the crowd can be used to search
• The principle is not new – anchor text used in “standard” search
• Don’t try to pass the Turing test?
![Page 16: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/16.jpg)
16Yahoo! Research
Anchor text
• When indexing a document D, include anchor text from links pointing to D.
www.ibm.com
Armonk, NY-based computergiant IBM announced today
Joe’s computer hardware linksCompaqHPIBM
Big Blue today announcedrecord profits for the quarter
![Page 17: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/17.jpg)
17Yahoo! Research
Challenges in tag-based search
• How do we use these tags for better search?
• How do you cope with spam?
• What’s the ratings and reputation system?
• The bigger challenge: where else can you exploit the power of the people?
• What are the incentive mechanisms?– Luis von Ahn (CMU): The ESP Game
![Page 18: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/18.jpg)
18Yahoo! Research
![Page 19: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/19.jpg)
19
More UGC: Social search
Indexing the knowledge in people’s heads
![Page 20: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/20.jpg)
20Yahoo! Research
![Page 21: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/21.jpg)
21Yahoo! Research
![Page 22: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/22.jpg)
22Yahoo! Research
Social content Social capital
![Page 23: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/23.jpg)
23Yahoo! Research
Incentives
![Page 24: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/24.jpg)
24Yahoo! Research
Incentives
• What assignment of incentives leads to good user behavior?
– What’s “good” user behavior?
– Good questions, good answers, new questions?
• Whom do you trust and why?
– Propagation of trust and distrust.
![Page 25: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/25.jpg)
25Yahoo! Research
Ratings and reputation
• Node reputation: Given a DAG with– a subset of nodes called GOOD
– another subset called BAD
– Find a measure of goodness for all other nodes.
• Node pair reputation: Given a DAG with a real-valued trust on the edges– Predict a real-valued trust for ordered node
pairs not joined by an edge
Metric labelling
![Page 26: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/26.jpg)
26
CPC advertisements
What pays the bills
![Page 27: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/27.jpg)
27Yahoo! Research
Ads
![Page 28: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/28.jpg)
28Yahoo! Research
Generic questions
• Of the various advertisers for a keyword, which one(s) get shown?
• What do they pay on a click through?
• The answers turn out to draw on insights from microeconomics
![Page 29: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/29.jpg)
29Yahoo! Research
Ads go in slots like this one
and this one.
![Page 30: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/30.jpg)
30Yahoo! Research
Advertisers generally prefer this slot
to this one
to this one
to this one.
![Page 31: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/31.jpg)
31Yahoo! Research
Click through rate r1 = 200 per hour
r2 = 150 per hour
r3 = 100 per hour
etc.
![Page 32: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/32.jpg)
32Yahoo! Research
Why did witbeckappliance win
over ristenbatt?
![Page 33: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/33.jpg)
33Yahoo! Research
First-cut assumption
• Click-through rate depends only on the slot, not on the advertisement
• In fact not true; more on this later.
![Page 34: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/34.jpg)
34Yahoo! Research
Advertiser’s value
• We assume that an advertiser j has a value vj per click through
– Some measure of downstream profit
• Say, click-through followed by• 96% of the time, no purchase
• 0.7% buy Dishwasher, profit $500
• 1.2% buy Vacuum Cleaner, profit $200
• 2.1% buy Cleaning agents, profit $1$ 5.921
![Page 35: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/35.jpg)
35Yahoo! Research
Example
• For the keyword miele, say an advertiser has a value of $10 per click.
• How much should he bid?
• How much should he be charged?
The value of a slot for an advertiser,
what he bids and
what he is charged, may all be different.
![Page 36: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/36.jpg)
36Yahoo! Research
Advertiser’s payoff in ad slot i
(Click-through rate) x (Value per click) – (Payment to search engine)
= ri vj – (Payment to Engine)
= ri vj – pij
Payment ofadvertiser j
in slot iFunction of all other bids.
![Page 37: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/37.jpg)
37Yahoo! Research
Two auction pricing mechanisms
• First price: The winner of the auction is the highest bidder, and pays his bid.
• Second price: The winner is the highest bidder, but pays the second-highest bid.
• Engine decides and announces pricing.
• What should an advertiser bid?
Not truthful.
![Page 38: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/38.jpg)
38Yahoo! Research
Second-price = Vickrey auction
• Consider first a single advt slot
• Winner pays the second-highest bid
• Vickrey: Truth-telling is a dominant strategy for each player (advertiser)
– No incentive to “game” or fake bids
![Page 39: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/39.jpg)
39Yahoo! Research
Auctions and pricing: multiple slots
• Overture’s model:
– Ads displayed in order of decreasing bid
– E.g., if advertiser A bids 10, B bids 2, C bids 4 – order ACB
• How do you price slots? Generalized Vickrey?
– Generalized second-price (GSP)
– Vickrey-Clark-Groves (VCG): each advertiser pays the externality he imposes on others
![Page 40: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/40.jpg)
40Yahoo! Research
Bidder A, $10
Bidder C, $4
Bidder B, $2
Pays 4
Pays 2
Generalized Second Price auction pricing
![Page 41: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/41.jpg)
41Yahoo! Research
VCG pricing
• Suppose click rates are 200 in the top slot, 100 in the second slot
• VCG payment of the second player (C) is 2 x 100 = 200
• For the first player, 4x(200-100) + 200
Externality on third player B.
Externality on C. Externality on B.
![Page 42: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/42.jpg)
42Yahoo! Research
VCG and GSP
• Truth-telling is a dominant strategy under VCG …
• Truth-telling not dominant under GSP!
Edelman, Ostrovsky, Schwarz
![Page 43: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/43.jpg)
43Yahoo! Research
VCG and GSP
• Static equilibrium of GSP is locally envy-free: no advertiser can improve his payoff by exchanging bids with advertiser in slot above.
• Depending on the mechanism, revenue varies: GSP ≥ VCG.
Edelman, Ostrovsky, Schwarz
Locally envy-free mechanisms correspondto Stable Marriage solutions.
![Page 44: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/44.jpg)
44Yahoo! Research
GSP for bid-ordering
• What’s good about bid-ordering and GSP?
–Advertisers like transparency
• What’s wrong with bid-ordering?
![Page 45: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/45.jpg)
45Yahoo! Research
Brand advertising?
Bid ordering(former Yahoo! order)
![Page 46: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/46.jpg)
46Yahoo! Research
![Page 47: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/47.jpg)
47Yahoo! Research
Revenue ordering
• Simplified version of Google’s ordering
– Each ad j has an expected click-through denoted CTRj
– Advertiser j’s bid is denoted bj
• Then, expected revenue from this advertiser is Rj = bj+1 x CTRj
• Order advertisers by Rj
– Payment by GSP
![Page 48: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/48.jpg)
48Yahoo! Research
![Page 49: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/49.jpg)
49Yahoo! Research
“current” Yahoo! ordering
![Page 50: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/50.jpg)
50Yahoo! Research
“Squashed” ordering
• Overture/Old Yahoo! scheme
– Order ads by bid
• Google (puportedly)
– Order by bid click-through rate (CTR)
• Squashing (Lahaie/Pennock)
• Key – advertisers react to mechanism!
s=0 s=1Order by bid*(CTR)sOverture Google?
![Page 51: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/51.jpg)
51Yahoo! Research
![Page 52: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/52.jpg)
52Yahoo! Research
Where do we go next?
• Premise:
– People don’t want to search
– People want to get tasks done
I want to book a vacation in Tuscany.Start Finish
![Page 53: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/53.jpg)
53Yahoo! Research
What is missing?
• Information integration
– Information extraction
– Schema normalization
• Mining social structure
– Tags, UGC
Welcome to The SavoyLocated on The Strand in the heartof the West End theatre district,
hotel near leicester squareSearch
![Page 54: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/54.jpg)
54Yahoo! Research
Computational microeconomics
• Reputation and incentive mechanisms
• Matching marketplaces– Jobs, dates, …
– Online matching everywhere• Hardest part is estimating the payoffs, not the
matching algorithm
• “Network effects”– Are 500 million users 500 times as valuable
as a million users? 5000 times?
![Page 55: 1 Web Search From Information Retrieval to Microeconomic Modeling Prabhakar Raghavan Yahoo! Research](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d4e5503460f94a2dbd1/html5/thumbnails/55.jpg)
55Yahoo! Research
A new convergence
• Monetization and economic value an intrinsic part of system design
– Not an afterthought
– Mistakes are costly!
• Computing meets humanities like never before – sociology, economics, anthropology …