SlideShare a Scribd company logo
1 of 23
What Questions are Worth Answering?
Ehren Reilly
Sr. Product Manager, Content,
Ask.com
Sentiment Analysis Innovation
Summit
San Francisco, CA
April 25, 2013
Overview
• Our challenge: What queries deserve an editorial
answer?
• Our approach to cost-effectively figuring this out
• Advantages of our approach
• Details of our approach
Our Challenge
• Ask.com Snapshot:
• Q&A service combining the power of search, quality
editorial content and content from the web
• Top 10 US internet site (according to comScore)
• 100 million unique users globally; 70 million unique
users in the US
• Search  Q&A
• When you come to Ask.com and ask a question, we give
you the answer to your question.
• Ask.com editors create answers to questions that are asked
frequently, based on search query data
• Problem: Not every query is suitable for evergreen, static
editorial answers
Our Challenge
What type of information does the query deserve?
• Entities & services (people, things, websites, products, media,
resources)
o General web search, shopping search, Wikipedia data,
tools, applications
• Dynamic data and frequently-changing facts (e.g., the
weather)
o Data partnerships
• Static, evergreen information, suitable for editorial expert
answers
o Writers, editors, crowd labor, etc. $$
• Extremely detailed/technical answer which needs long article
by a true expert
Our Challenge
Not answer requests Wants dynamic data
answer
Wants evergreen expert
answer
• Facebook login
• Barack Obama
• Tickets Seattle to
Miami
• Olympus Has Fallen
• Sonicare HX6711/02
• Selena Gomez photos
• Philippines map
• German Shepherds
• Chichen Itza timelapse
• Salary calculator
• What time is it in
Bangkok
• Dollars to pounds
• SF Giants score
• Weather in Cleveland
• What’s my IP address
• Kim Kardashian
pregnant?
• NBA assists leader
• Where is Justin
Bieber?
• Oldest living person
• Gay marriage states
• When was the Ming
Dynasty
• Tom Cruise baby name
• How long to bake chicken
• What is Renaissance art
• Highest alcohol beer
• Head gasket repair cost
• Abraham Lincoln’s wife
• Parachute material
• Most reliable dishwashers
• How to remove hair dye
What type of information does the query want?
Our Challenge
Editorial Answerability Spectrum
Navigational
Dynamic
FactsEntitiesShopping
Evergreen
Facts
How do we pick out these answerable
evergreen fact queries?
• Let the editors do it themselves?
o Valuable editorial time wasted considering obvious stuff
o For crowd editorial labor, conflict of interest  “OK” bias
• Crowd labor vetting?
o Hard to communicate task
o Still very costly
• Template-based filters?
o Coverage is too low
o Lots of work to develop these
• Machine learning?
o Very fuzzy problem
o Target set is a small segment of huge search space
o Hard to achieve high accuracy
Our Hybrid Approach
1. Filter out the obvious stuff (e.g., “Facebook.com”, “What time is
it”, “What does „looking a gift horse in the mouth‟ mean?”)
2. Dedicated classifiers to filter out specific types of non-suitable
queries
• Duplicates & near-duplicates
• Navigational
• Adult / profane / creepy
• Temporal / dynamic / timely
• Shopping / product search
• Wiki / entity exact match
3. Build machine learning “answerability” model for the tricky
remaining cases
4. Where the model returns low confidence, send those queries to
crowd labor for classification
Advantages to This Approach
Evergreen
Facts
Advantages to This Approach
Evergreen
Facts
Advantages to This Approach
Evergreen
Facts
Advantages to This Approach
Evergreen
Facts
Don‟t
Send to
Editorial
Advantages to This Approach
Evergreen
Facts
Don‟t
Send to
Editorial
Requires Human
Review
Advantages to This Approach
• Filtering and partial automation first makes human review
much less costly
o Tasks requiring human scoring reduced by 97%
• Domain of ML model is narrower than entire query mix, which
improves accuracy
• Making the model better over time
o Human rating data becomes training data for algorithm
o Gradually, algorithm gets better, you need fewer human
ratings
Human Rater Biases
• Two very different tasks:
o Look for attribute X, which occurs in 1% of data.
o Look for attribute X, which occurs in 50% of data.
• The harder you have to look for instances of X, the more things
start to look like X.
o Your sensitivity increases. You get trigger-happy.
Human Rater Biases
Thought experiment:
“Listen for any naughty words or phrases”
 Corpus 1: Nationally televised sports color commentary
 Corpus 2: Gangster rap music
• Some words sound bad in the nationally televised sports
context, but wouldn‟t in the gangster rap context.
• Cognitive psychologists call this the Contrast Effect.
Human Rater Biases
• We gave two sets of crowdsource workers (same agency,
same pay rate) the same data, mixed in with two different
surrounding data sets
o Group A: Raw query file
o Group B: Filtered with heuristics and templates first
• Of the queries that group A thought were answerable, Group
B only though 64% of those were answerable
• Queries where the two groups disagreed where
overwhelmingly false positives by Group A, rather than false
negatives from Group B:
• how you spell a word
• how much does a book of stamps cost
• is randy fenoli married
• when does the alabama football game start
• where to donate old magazines
What Crowdsource Writers Will and Won‟t
Do for You
• Don‟t rely on crowdsource workers to self-select which
tasks are viable
o “Only answer the answerable queries” (and we only pay
you for what you answer)
o Writers biased towards everything being answerable
• Exception: If the task is too big, they are happy to flag those
o How to repair a transmission
o History of China
o US senators all time
o How does organic chemistry work
Easy Filters: Dynamic
Easy Filters: Dumb
What to Include in Training Data
• Some question patterns are almost universally answerable
questions
o Who invented [NP]?
o Where was [person] born?
o How to [cooking verb] a [food item]
o What does […] mean?
• We grab these queries using template filters, and don‟t need ML
• Should we included these in our training data?
• This is an empirical question. Does the algorithm perform better
or worse if the “easy” data is included in the training data?
• In this specific case, the model is more accurate when trained
without “easy” data
Conclusions
• If you have a firehose of data, don‟t just:
o Send it to crowdsourcers
o Try to build a ML model
• Instead, figure out what the “easy” cases are, and deal with
those separately, using common sense rules
• Put your crowdsourcing and machine learning efforts on just
the hard part of the problem
THANK YOU
Ehren Reilly
ehren.reilly@ask.com
@ehrenreilly

More Related Content

Similar to What Questions Are Worth Answering?

Engaging with Users on Public Social Media
Engaging with Users on Public Social MediaEngaging with Users on Public Social Media
Engaging with Users on Public Social MediaJeffrey Nichols
 
Pubcon Vegas 2010 - Social Media: Measurements & Tools
Pubcon Vegas 2010 - Social Media: Measurements & ToolsPubcon Vegas 2010 - Social Media: Measurements & Tools
Pubcon Vegas 2010 - Social Media: Measurements & ToolsAdam Proehl
 
Human computation, crowdsourcing and social: An industrial perspective
Human computation, crowdsourcing and social: An industrial perspectiveHuman computation, crowdsourcing and social: An industrial perspective
Human computation, crowdsourcing and social: An industrial perspectiveoralonso
 
Girl Develop It: Intro to Content Strategy - May 2017
Girl Develop It: Intro to Content Strategy - May 2017Girl Develop It: Intro to Content Strategy - May 2017
Girl Develop It: Intro to Content Strategy - May 2017David Dylan Thomas
 
2016 03-15 lean @ lsc rotterdam by Tristan Kromer
2016 03-15 lean @ lsc rotterdam by Tristan Kromer2016 03-15 lean @ lsc rotterdam by Tristan Kromer
2016 03-15 lean @ lsc rotterdam by Tristan KromerFrançois la Rivière
 
The Hive Think Tank: Machine Learning at Pinterest by Jure Leskovec
The Hive Think Tank: Machine Learning at Pinterest by Jure LeskovecThe Hive Think Tank: Machine Learning at Pinterest by Jure Leskovec
The Hive Think Tank: Machine Learning at Pinterest by Jure LeskovecThe Hive
 
LIB300 Week 9 finding, analyzing, and documenting information
LIB300 Week 9 finding, analyzing, and documenting informationLIB300 Week 9 finding, analyzing, and documenting information
LIB300 Week 9 finding, analyzing, and documenting informationDr. Russell Rodrigo
 
2015 Find You - Fuzzy Big Data - A Case Study
2015 Find You - Fuzzy Big Data - A Case Study2015 Find You - Fuzzy Big Data - A Case Study
2015 Find You - Fuzzy Big Data - A Case StudyPhilip Topham
 
Culture & km survey.aqs.rev.10.8.14
Culture & km survey.aqs.rev.10.8.14Culture & km survey.aqs.rev.10.8.14
Culture & km survey.aqs.rev.10.8.14ronsasiela
 
Reactive Writing Techniques for Rewarding and Retaining Users
Reactive Writing Techniques for Rewarding and Retaining UsersReactive Writing Techniques for Rewarding and Retaining Users
Reactive Writing Techniques for Rewarding and Retaining Usersgrebstock
 
Management by data
Management by dataManagement by data
Management by dataLuca Foresti
 
Content Strategy for Business: What You Need to Know
Content Strategy for Business: What You Need to KnowContent Strategy for Business: What You Need to Know
Content Strategy for Business: What You Need to KnowDavid Dylan Thomas
 
Introduction to Content Strategy - NetSquared Remix
Introduction to Content Strategy - NetSquared RemixIntroduction to Content Strategy - NetSquared Remix
Introduction to Content Strategy - NetSquared RemixDavid Dylan Thomas
 
So You Wanna Be a Content Strategist: SXSW 2016
So You Wanna Be a Content Strategist: SXSW 2016So You Wanna Be a Content Strategist: SXSW 2016
So You Wanna Be a Content Strategist: SXSW 2016David Dylan Thomas
 
Data Con LA 2022 - Real world consumer segmentation
Data Con LA 2022 - Real world consumer segmentationData Con LA 2022 - Real world consumer segmentation
Data Con LA 2022 - Real world consumer segmentationData Con LA
 

Similar to What Questions Are Worth Answering? (20)

Engaging with Users on Public Social Media
Engaging with Users on Public Social MediaEngaging with Users on Public Social Media
Engaging with Users on Public Social Media
 
What Is Content Strategy?
What Is Content Strategy?What Is Content Strategy?
What Is Content Strategy?
 
Pubcon Vegas 2010 - Social Media: Measurements & Tools
Pubcon Vegas 2010 - Social Media: Measurements & ToolsPubcon Vegas 2010 - Social Media: Measurements & Tools
Pubcon Vegas 2010 - Social Media: Measurements & Tools
 
Human computation, crowdsourcing and social: An industrial perspective
Human computation, crowdsourcing and social: An industrial perspectiveHuman computation, crowdsourcing and social: An industrial perspective
Human computation, crowdsourcing and social: An industrial perspective
 
Girl Develop It: Intro to Content Strategy - May 2017
Girl Develop It: Intro to Content Strategy - May 2017Girl Develop It: Intro to Content Strategy - May 2017
Girl Develop It: Intro to Content Strategy - May 2017
 
Data Mining Lecture_2.pptx
Data Mining Lecture_2.pptxData Mining Lecture_2.pptx
Data Mining Lecture_2.pptx
 
2016 03-15 lean @ lsc rotterdam by Tristan Kromer
2016 03-15 lean @ lsc rotterdam by Tristan Kromer2016 03-15 lean @ lsc rotterdam by Tristan Kromer
2016 03-15 lean @ lsc rotterdam by Tristan Kromer
 
The Hive Think Tank: Machine Learning at Pinterest by Jure Leskovec
The Hive Think Tank: Machine Learning at Pinterest by Jure LeskovecThe Hive Think Tank: Machine Learning at Pinterest by Jure Leskovec
The Hive Think Tank: Machine Learning at Pinterest by Jure Leskovec
 
LIB300 Week 9 finding, analyzing, and documenting information
LIB300 Week 9 finding, analyzing, and documenting informationLIB300 Week 9 finding, analyzing, and documenting information
LIB300 Week 9 finding, analyzing, and documenting information
 
2015 Find You - Fuzzy Big Data - A Case Study
2015 Find You - Fuzzy Big Data - A Case Study2015 Find You - Fuzzy Big Data - A Case Study
2015 Find You - Fuzzy Big Data - A Case Study
 
HCI_Lecture04.pptx
HCI_Lecture04.pptxHCI_Lecture04.pptx
HCI_Lecture04.pptx
 
Culture & km survey.aqs.rev.10.8.14
Culture & km survey.aqs.rev.10.8.14Culture & km survey.aqs.rev.10.8.14
Culture & km survey.aqs.rev.10.8.14
 
Reactive Writing Techniques for Rewarding and Retaining Users
Reactive Writing Techniques for Rewarding and Retaining UsersReactive Writing Techniques for Rewarding and Retaining Users
Reactive Writing Techniques for Rewarding and Retaining Users
 
Searching skills
Searching skillsSearching skills
Searching skills
 
Management by data
Management by dataManagement by data
Management by data
 
Content Strategy for Business: What You Need to Know
Content Strategy for Business: What You Need to KnowContent Strategy for Business: What You Need to Know
Content Strategy for Business: What You Need to Know
 
ARlab RESEARCH | Social search
ARlab RESEARCH | Social searchARlab RESEARCH | Social search
ARlab RESEARCH | Social search
 
Introduction to Content Strategy - NetSquared Remix
Introduction to Content Strategy - NetSquared RemixIntroduction to Content Strategy - NetSquared Remix
Introduction to Content Strategy - NetSquared Remix
 
So You Wanna Be a Content Strategist: SXSW 2016
So You Wanna Be a Content Strategist: SXSW 2016So You Wanna Be a Content Strategist: SXSW 2016
So You Wanna Be a Content Strategist: SXSW 2016
 
Data Con LA 2022 - Real world consumer segmentation
Data Con LA 2022 - Real world consumer segmentationData Con LA 2022 - Real world consumer segmentation
Data Con LA 2022 - Real world consumer segmentation
 

Recently uploaded

Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...Neo4j
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Enterprise Knowledge
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024The Digital Insurer
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Drew Madelung
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 

Recently uploaded (20)

Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 

What Questions Are Worth Answering?

  • 1. What Questions are Worth Answering? Ehren Reilly Sr. Product Manager, Content, Ask.com Sentiment Analysis Innovation Summit San Francisco, CA April 25, 2013
  • 2. Overview • Our challenge: What queries deserve an editorial answer? • Our approach to cost-effectively figuring this out • Advantages of our approach • Details of our approach
  • 3. Our Challenge • Ask.com Snapshot: • Q&A service combining the power of search, quality editorial content and content from the web • Top 10 US internet site (according to comScore) • 100 million unique users globally; 70 million unique users in the US • Search  Q&A • When you come to Ask.com and ask a question, we give you the answer to your question. • Ask.com editors create answers to questions that are asked frequently, based on search query data • Problem: Not every query is suitable for evergreen, static editorial answers
  • 4. Our Challenge What type of information does the query deserve? • Entities & services (people, things, websites, products, media, resources) o General web search, shopping search, Wikipedia data, tools, applications • Dynamic data and frequently-changing facts (e.g., the weather) o Data partnerships • Static, evergreen information, suitable for editorial expert answers o Writers, editors, crowd labor, etc. $$ • Extremely detailed/technical answer which needs long article by a true expert
  • 5. Our Challenge Not answer requests Wants dynamic data answer Wants evergreen expert answer • Facebook login • Barack Obama • Tickets Seattle to Miami • Olympus Has Fallen • Sonicare HX6711/02 • Selena Gomez photos • Philippines map • German Shepherds • Chichen Itza timelapse • Salary calculator • What time is it in Bangkok • Dollars to pounds • SF Giants score • Weather in Cleveland • What’s my IP address • Kim Kardashian pregnant? • NBA assists leader • Where is Justin Bieber? • Oldest living person • Gay marriage states • When was the Ming Dynasty • Tom Cruise baby name • How long to bake chicken • What is Renaissance art • Highest alcohol beer • Head gasket repair cost • Abraham Lincoln’s wife • Parachute material • Most reliable dishwashers • How to remove hair dye What type of information does the query want?
  • 6. Our Challenge Editorial Answerability Spectrum Navigational Dynamic FactsEntitiesShopping Evergreen Facts
  • 7. How do we pick out these answerable evergreen fact queries? • Let the editors do it themselves? o Valuable editorial time wasted considering obvious stuff o For crowd editorial labor, conflict of interest  “OK” bias • Crowd labor vetting? o Hard to communicate task o Still very costly • Template-based filters? o Coverage is too low o Lots of work to develop these • Machine learning? o Very fuzzy problem o Target set is a small segment of huge search space o Hard to achieve high accuracy
  • 8. Our Hybrid Approach 1. Filter out the obvious stuff (e.g., “Facebook.com”, “What time is it”, “What does „looking a gift horse in the mouth‟ mean?”) 2. Dedicated classifiers to filter out specific types of non-suitable queries • Duplicates & near-duplicates • Navigational • Adult / profane / creepy • Temporal / dynamic / timely • Shopping / product search • Wiki / entity exact match 3. Build machine learning “answerability” model for the tricky remaining cases 4. Where the model returns low confidence, send those queries to crowd labor for classification
  • 9. Advantages to This Approach Evergreen Facts
  • 10. Advantages to This Approach Evergreen Facts
  • 11. Advantages to This Approach Evergreen Facts
  • 12. Advantages to This Approach Evergreen Facts Don‟t Send to Editorial
  • 13. Advantages to This Approach Evergreen Facts Don‟t Send to Editorial Requires Human Review
  • 14. Advantages to This Approach • Filtering and partial automation first makes human review much less costly o Tasks requiring human scoring reduced by 97% • Domain of ML model is narrower than entire query mix, which improves accuracy • Making the model better over time o Human rating data becomes training data for algorithm o Gradually, algorithm gets better, you need fewer human ratings
  • 15. Human Rater Biases • Two very different tasks: o Look for attribute X, which occurs in 1% of data. o Look for attribute X, which occurs in 50% of data. • The harder you have to look for instances of X, the more things start to look like X. o Your sensitivity increases. You get trigger-happy.
  • 16. Human Rater Biases Thought experiment: “Listen for any naughty words or phrases”  Corpus 1: Nationally televised sports color commentary  Corpus 2: Gangster rap music • Some words sound bad in the nationally televised sports context, but wouldn‟t in the gangster rap context. • Cognitive psychologists call this the Contrast Effect.
  • 17. Human Rater Biases • We gave two sets of crowdsource workers (same agency, same pay rate) the same data, mixed in with two different surrounding data sets o Group A: Raw query file o Group B: Filtered with heuristics and templates first • Of the queries that group A thought were answerable, Group B only though 64% of those were answerable • Queries where the two groups disagreed where overwhelmingly false positives by Group A, rather than false negatives from Group B: • how you spell a word • how much does a book of stamps cost • is randy fenoli married • when does the alabama football game start • where to donate old magazines
  • 18. What Crowdsource Writers Will and Won‟t Do for You • Don‟t rely on crowdsource workers to self-select which tasks are viable o “Only answer the answerable queries” (and we only pay you for what you answer) o Writers biased towards everything being answerable • Exception: If the task is too big, they are happy to flag those o How to repair a transmission o History of China o US senators all time o How does organic chemistry work
  • 21. What to Include in Training Data • Some question patterns are almost universally answerable questions o Who invented [NP]? o Where was [person] born? o How to [cooking verb] a [food item] o What does […] mean? • We grab these queries using template filters, and don‟t need ML • Should we included these in our training data? • This is an empirical question. Does the algorithm perform better or worse if the “easy” data is included in the training data? • In this specific case, the model is more accurate when trained without “easy” data
  • 22. Conclusions • If you have a firehose of data, don‟t just: o Send it to crowdsourcers o Try to build a ML model • Instead, figure out what the “easy” cases are, and deal with those separately, using common sense rules • Put your crowdsourcing and machine learning efforts on just the hard part of the problem