SlideShare une entreprise Scribd logo
1  sur  32
Télécharger pour lire hors ligne
1
Diversifying
Autocomplete
Felipe Besson
Haystack / MICES / Berlin Buzzwords
June 11, 2020
2
20+ classifieds brands worldwide
350+ mi users each month
5000+ employees
35 offices
Online Classifieds Platform
25+ countries
OLX Europe Discovery Cycle
3
Recommendations
home page search page
ad page
All steps are interconnected!
4
● Users have different intents
● What can break the dialogue with the user ?
○ Broad queries (Autocomplete and Search)
○ Ambiguity (Query understanding)
○ Bad Interactions (Recommendations)
Diversifying search results
5
Strength the dialogue with the user
● Dealing with broad queries
○ Autocomplete
○ Search
● Item showcase for new or exploring users
● Gathering more interactions to improve recommendations
● Autocomplete
What will be covered and how ?
6
● Broad queries problem in autocomplete
● Techniques to promote diversification
● Our use case:
○ Autocomplete at OLX Europe
What is Autocomplete ?
7
A tool to talk directly to the user
● Guide users to good queries
● Help query understanding to understand
● Fast response/reaction
● Help tackling search relevance earlier as possible
Autocomplete at OLX Europe
8
● Suggest popular searches with category filters
● Covers 7 different countries
● > 50 mi requests per day
● Responsible for 40% of total searches
● Ranks suggestions by popularity and narrowness
○ but ...
Broad query problem ...
9
What is my intent ?
What if I don't know any Vespa
model ?
popularity
What if I have a Vespa and
want some accessory ?
Broad query effect ...
Fashion
Bags and
accessories
Footwear
Clothing
Watches and
Jewelry
Notions
Other bags and
accessories
Woman
Sunglasses
Man
Woman
Man
Watches
Jewelry
10
Different topics
Level 1 (L1)
Level 2 (L2)
Level 3 (L3)
Gucci
Wallets
Handbags
Health and
Beauty
Perfumes
Medical care
Autocomplete
suggestions
Breaks in the dialogue with user
11
● We jumped to premature conclusions
○ Show very specific popular suggestions (Vespa models)
● We could have asked more
○ Show more possibilities (like accessories)
● Maybe we will never have the chance to ask more
○ Popularity feedback loop ("rich get richer")
Diversifying autocomplete suggestions
12
Improve user experience on broad queries
● Minimize overspecialization of suggestions
● Give an overview of different available item categories
● Break popularity feedback loop
● Refine the query (user intents)
The goal
13
Diversifying autocomplete category suggestions for broad queries
Broad queries =
popular queries
AND contain categories with many search results
AND those categories are not yet suggested!
How to apply diversification ?
14
Inspiration from Web Search and Information retrieval
Explicit diversification
○ From query (information needs)
○ Increase Coverage
○ Broad queries
Based on Search result diversification: http://www.dcs.gla.ac.uk/~craigm/publications/santos2015ftir.pdf
How can we measure coverage ?
15
Step 1: Clustering documents into topics
○ Facets, categories, colors, word embeddings, ...
891
...
36
...
37
...
903 3
topics
topics
probability
How can we measure coverage ?
16
Step 2: Measure dispersion of topics distribution
GINI Coefficient: https://opensourceconnections.com/blog/2019/09/05/diversity-vs-relevance
<>
GINI Coefficient
Shannon Entropy
topicstopics
probability
probability
Shannon Entropy
17
Measures level of information in a probability distribution
A B C
High Knowledge Medium Knowledge Low Knowledge
Low Surprise Medium Surprise High Surprise
entropy = 0 entropy = 0.81 entropy = 1.5
Shannon Entropy for e-commerces
18
1. Cluster document into categories (or any other criteria)
2. Category probability
entropy: 2.38 entropy: 0.52
Entropy from another perspective
19Extracted from: https://medium.com/udacity/shannon-entropy-information-gain-and-picking-balls-from-buckets-5810d35d54b4
On average, how many questions do we need to ask to find out what letter it is?
Entropy = 0
Bucket 1
Entropy = 1.75
Bucket 2
Entropy = 2.0
Bucket 3
Akinator: https://en.wikipedia.org/wiki/Akinator
Entropy from another perspective
20
Extracted from: https://medium.com/udacity/shannon-entropy-information-gain-and-picking-balls-from-buckets-5810d35d54b4
Bucket 3 (2 questions on overage)
Bucket 2 (1.75 questions on average)
Coming back to the autocomplete
21
On average, how many questions can we ask to make sure we cover all user intents ?
each suggestion we give = a different question we make
○ 0 questions for very specific queries (low entropy)
○ n questions for broad queries
■ How many is n ?
■ How can we define these questions ?
How many questions can we ask ?
22
possible question!
entropy of each category
10 slots
Entropy = # of different questions
Maximum diversity is 10 different suggestions!
● Each category has p(x) = 0.1 and e(x) = 0.33
How to pick each suggestion ?
23
0.33
too few results Narrow queries
candidates
Generation new suggestions
Fashion
Bags and
accessories
Footwear Clothing
Watches and
Jewelry
Notions
24
Gucci
Health and
Beauty
H(X) = 1.27
p(x) = 0.56
e(x) = 0.47
p(x) = 0.15
e(x) = 0.41
p(x) = 0.14
e(x) = 0.39
p(x) = 0.09
e(x) = 0.32
p(x) = 0.05
e(x) = 0.22
p(x) = 0.002
e(x) = 0.019
L2
Experiment pipeline
25
Goal: Expand suggestions for broad queries
Expansion example
26
Gucci
Before After
inherited popularity
Expansion example
27
iphone
Before After
Experiment Scope
28
● 2 countries (C1 and C2)
● Expansions for less than 5% of suggested queries but covered:
○ 26% of total searches for C1
○ 17% of total searches for C2
● Compared the performance of both groups
○ broad queries: expanded vs not expanded
Primary metrics Description C1 C2
suggest_search_rate Autocomplete usage: # suggested searches / # total searches +10.41% +0.72%
pos_filter_rate Search filters applied after picking expanded suggestions -3.14% -5.14%
Experiment Results
29
● Diversification impacted user behaviour in autocomplete
● C1 users interacted more with autocomplete suggestions
● Did C2 users pick less suggestions but better ones ?
Experiment Results
30
Query metrics* Description C1 C2
suggest_ctr Uplift in ad clicks from expanded query +3.64% -3.86
suggest_reply_rate Uplift in ad replies from expanded query +1.81% +0.26%
Suggestion metrics* Description C1 C2
suggest_cat_ctr Uplift in ad clicks from expanded suggestions (category) +2.24% +9.48%
suggest_cat_reply_rate Uplift in ad replies from expanded suggestions (category) +6.13% +13.01%
● Promising for C1 users in general
● In C2, we might have replaced relevant suggestions
● In both countries, new suggested categories look relevant
Considerations and Future
31
● Early stage: first and simple iteration
● Extend experiment
○ Affect more queries and add more countries
● Impact short vs long term
○ Consider rank (top n results)
○ Explore more clustering dimensions
○ Define entropy and popularity thresholds (prior and observed)
Thanks
32
linkedin.com/in/felipe-besson
@fmbesson

Contenu connexe

Similaire à Diversifying Autocomplete

How to Succeed as a PM by Native Instruments fmr Dir of Product
How to Succeed as a PM by Native Instruments fmr Dir of ProductHow to Succeed as a PM by Native Instruments fmr Dir of Product
How to Succeed as a PM by Native Instruments fmr Dir of ProductProduct School
 
Vincent Boom (Shoulders of Giants): How To Use Your Community To Transform Yo...
Vincent Boom (Shoulders of Giants): How To Use Your Community To Transform Yo...Vincent Boom (Shoulders of Giants): How To Use Your Community To Transform Yo...
Vincent Boom (Shoulders of Giants): How To Use Your Community To Transform Yo...FeverBee Limited
 
Top 10 AdWords Tactics for B2B Lead Gen Campaigns
Top 10 AdWords Tactics for B2B Lead Gen CampaignsTop 10 AdWords Tactics for B2B Lead Gen Campaigns
Top 10 AdWords Tactics for B2B Lead Gen CampaignsSwydo
 
Leverage The Power of Small Data
Leverage The Power of Small DataLeverage The Power of Small Data
Leverage The Power of Small DataKaryn Zuidinga
 
Final Quiz – 100 QuestionsTop of FormOne of Wal-Marts adverti.docx
Final Quiz – 100 QuestionsTop of FormOne of Wal-Marts adverti.docxFinal Quiz – 100 QuestionsTop of FormOne of Wal-Marts adverti.docx
Final Quiz – 100 QuestionsTop of FormOne of Wal-Marts adverti.docxvoversbyobersby
 
Storytelling with Data (Global Engagement Summit at Northwestern University 2...
Storytelling with Data (Global Engagement Summit at Northwestern University 2...Storytelling with Data (Global Engagement Summit at Northwestern University 2...
Storytelling with Data (Global Engagement Summit at Northwestern University 2...Sara Hooker
 
Competion-Pack 2023.pdf
Competion-Pack 2023.pdfCompetion-Pack 2023.pdf
Competion-Pack 2023.pdfAimeMoh
 
Competion-Pack 2023.pdf
Competion-Pack 2023.pdfCompetion-Pack 2023.pdf
Competion-Pack 2023.pdfAimeMoh
 
Heuristic website review: It's your website. But it's your members' experience.
Heuristic website review: It's your website. But it's your members' experience.Heuristic website review: It's your website. But it's your members' experience.
Heuristic website review: It's your website. But it's your members' experience.Ray van Hilst
 
Philanthropitch Austin Application Q&A Session
Philanthropitch Austin Application Q&A SessionPhilanthropitch Austin Application Q&A Session
Philanthropitch Austin Application Q&A SessionRoyal Kelly
 
Digital analytics lecture1
Digital analytics lecture1Digital analytics lecture1
Digital analytics lecture1Joni Salminen
 
Good Questions, Good Products: 31+ Questions for Product Makers and Managers
Good Questions, Good Products: 31+ Questions for Product Makers and ManagersGood Questions, Good Products: 31+ Questions for Product Makers and Managers
Good Questions, Good Products: 31+ Questions for Product Makers and ManagersEli Holder
 
Making Multi Million Dollar from WordPress Plugin without Paid Marketing
Making Multi Million Dollar from WordPress Plugin without Paid MarketingMaking Multi Million Dollar from WordPress Plugin without Paid Marketing
Making Multi Million Dollar from WordPress Plugin without Paid MarketingMd Nizam Uddin
 
Aligning Product Strategy with Customer Feature Requests
Aligning Product Strategy with Customer Feature RequestsAligning Product Strategy with Customer Feature Requests
Aligning Product Strategy with Customer Feature RequestsProductPlan
 
Marketing Strategy for B2B
Marketing Strategy for B2BMarketing Strategy for B2B
Marketing Strategy for B2BAmrit Sagar
 
How to optimise your social media campaigns
How to optimise your social media campaignsHow to optimise your social media campaigns
How to optimise your social media campaignsDigital Visitor
 
Feedback & Surveys - How to use the Constant Contact Toolkit Part 2
Feedback & Surveys - How to use the Constant Contact Toolkit Part 2Feedback & Surveys - How to use the Constant Contact Toolkit Part 2
Feedback & Surveys - How to use the Constant Contact Toolkit Part 2Frithjof Petscheleit
 
User Research 101: DIY Quick Course - CodeMash 2.0.1.1.
User Research 101: DIY Quick Course - CodeMash 2.0.1.1.User Research 101: DIY Quick Course - CodeMash 2.0.1.1.
User Research 101: DIY Quick Course - CodeMash 2.0.1.1.Carol Smith
 
Opinion Driven Decision Support System
Opinion Driven Decision Support SystemOpinion Driven Decision Support System
Opinion Driven Decision Support SystemKavita Ganesan
 
Generating Quality Hypotheses For Higher Uplifts | Masters of Conversion by VWO
Generating Quality Hypotheses For Higher Uplifts | Masters of Conversion by VWOGenerating Quality Hypotheses For Higher Uplifts | Masters of Conversion by VWO
Generating Quality Hypotheses For Higher Uplifts | Masters of Conversion by VWOVWO
 

Similaire à Diversifying Autocomplete (20)

How to Succeed as a PM by Native Instruments fmr Dir of Product
How to Succeed as a PM by Native Instruments fmr Dir of ProductHow to Succeed as a PM by Native Instruments fmr Dir of Product
How to Succeed as a PM by Native Instruments fmr Dir of Product
 
Vincent Boom (Shoulders of Giants): How To Use Your Community To Transform Yo...
Vincent Boom (Shoulders of Giants): How To Use Your Community To Transform Yo...Vincent Boom (Shoulders of Giants): How To Use Your Community To Transform Yo...
Vincent Boom (Shoulders of Giants): How To Use Your Community To Transform Yo...
 
Top 10 AdWords Tactics for B2B Lead Gen Campaigns
Top 10 AdWords Tactics for B2B Lead Gen CampaignsTop 10 AdWords Tactics for B2B Lead Gen Campaigns
Top 10 AdWords Tactics for B2B Lead Gen Campaigns
 
Leverage The Power of Small Data
Leverage The Power of Small DataLeverage The Power of Small Data
Leverage The Power of Small Data
 
Final Quiz – 100 QuestionsTop of FormOne of Wal-Marts adverti.docx
Final Quiz – 100 QuestionsTop of FormOne of Wal-Marts adverti.docxFinal Quiz – 100 QuestionsTop of FormOne of Wal-Marts adverti.docx
Final Quiz – 100 QuestionsTop of FormOne of Wal-Marts adverti.docx
 
Storytelling with Data (Global Engagement Summit at Northwestern University 2...
Storytelling with Data (Global Engagement Summit at Northwestern University 2...Storytelling with Data (Global Engagement Summit at Northwestern University 2...
Storytelling with Data (Global Engagement Summit at Northwestern University 2...
 
Competion-Pack 2023.pdf
Competion-Pack 2023.pdfCompetion-Pack 2023.pdf
Competion-Pack 2023.pdf
 
Competion-Pack 2023.pdf
Competion-Pack 2023.pdfCompetion-Pack 2023.pdf
Competion-Pack 2023.pdf
 
Heuristic website review: It's your website. But it's your members' experience.
Heuristic website review: It's your website. But it's your members' experience.Heuristic website review: It's your website. But it's your members' experience.
Heuristic website review: It's your website. But it's your members' experience.
 
Philanthropitch Austin Application Q&A Session
Philanthropitch Austin Application Q&A SessionPhilanthropitch Austin Application Q&A Session
Philanthropitch Austin Application Q&A Session
 
Digital analytics lecture1
Digital analytics lecture1Digital analytics lecture1
Digital analytics lecture1
 
Good Questions, Good Products: 31+ Questions for Product Makers and Managers
Good Questions, Good Products: 31+ Questions for Product Makers and ManagersGood Questions, Good Products: 31+ Questions for Product Makers and Managers
Good Questions, Good Products: 31+ Questions for Product Makers and Managers
 
Making Multi Million Dollar from WordPress Plugin without Paid Marketing
Making Multi Million Dollar from WordPress Plugin without Paid MarketingMaking Multi Million Dollar from WordPress Plugin without Paid Marketing
Making Multi Million Dollar from WordPress Plugin without Paid Marketing
 
Aligning Product Strategy with Customer Feature Requests
Aligning Product Strategy with Customer Feature RequestsAligning Product Strategy with Customer Feature Requests
Aligning Product Strategy with Customer Feature Requests
 
Marketing Strategy for B2B
Marketing Strategy for B2BMarketing Strategy for B2B
Marketing Strategy for B2B
 
How to optimise your social media campaigns
How to optimise your social media campaignsHow to optimise your social media campaigns
How to optimise your social media campaigns
 
Feedback & Surveys - How to use the Constant Contact Toolkit Part 2
Feedback & Surveys - How to use the Constant Contact Toolkit Part 2Feedback & Surveys - How to use the Constant Contact Toolkit Part 2
Feedback & Surveys - How to use the Constant Contact Toolkit Part 2
 
User Research 101: DIY Quick Course - CodeMash 2.0.1.1.
User Research 101: DIY Quick Course - CodeMash 2.0.1.1.User Research 101: DIY Quick Course - CodeMash 2.0.1.1.
User Research 101: DIY Quick Course - CodeMash 2.0.1.1.
 
Opinion Driven Decision Support System
Opinion Driven Decision Support SystemOpinion Driven Decision Support System
Opinion Driven Decision Support System
 
Generating Quality Hypotheses For Higher Uplifts | Masters of Conversion by VWO
Generating Quality Hypotheses For Higher Uplifts | Masters of Conversion by VWOGenerating Quality Hypotheses For Higher Uplifts | Masters of Conversion by VWO
Generating Quality Hypotheses For Higher Uplifts | Masters of Conversion by VWO
 

Dernier

W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...panagenda
 
Love witchcraft +27768521739 Binding love spell in Sandy Springs, GA |psychic...
Love witchcraft +27768521739 Binding love spell in Sandy Springs, GA |psychic...Love witchcraft +27768521739 Binding love spell in Sandy Springs, GA |psychic...
Love witchcraft +27768521739 Binding love spell in Sandy Springs, GA |psychic...chiefasafspells
 
Large-scale Logging Made Easy: Meetup at Deutsche Bank 2024
Large-scale Logging Made Easy: Meetup at Deutsche Bank 2024Large-scale Logging Made Easy: Meetup at Deutsche Bank 2024
Large-scale Logging Made Easy: Meetup at Deutsche Bank 2024VictoriaMetrics
 
%+27788225528 love spells in Knoxville Psychic Readings, Attraction spells,Br...
%+27788225528 love spells in Knoxville Psychic Readings, Attraction spells,Br...%+27788225528 love spells in Knoxville Psychic Readings, Attraction spells,Br...
%+27788225528 love spells in Knoxville Psychic Readings, Attraction spells,Br...masabamasaba
 
What Goes Wrong with Language Definitions and How to Improve the Situation
What Goes Wrong with Language Definitions and How to Improve the SituationWhat Goes Wrong with Language Definitions and How to Improve the Situation
What Goes Wrong with Language Definitions and How to Improve the SituationJuha-Pekka Tolvanen
 
%+27788225528 love spells in Huntington Beach Psychic Readings, Attraction sp...
%+27788225528 love spells in Huntington Beach Psychic Readings, Attraction sp...%+27788225528 love spells in Huntington Beach Psychic Readings, Attraction sp...
%+27788225528 love spells in Huntington Beach Psychic Readings, Attraction sp...masabamasaba
 
%in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park %in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park masabamasaba
 
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisamasabamasaba
 
%+27788225528 love spells in Toronto Psychic Readings, Attraction spells,Brin...
%+27788225528 love spells in Toronto Psychic Readings, Attraction spells,Brin...%+27788225528 love spells in Toronto Psychic Readings, Attraction spells,Brin...
%+27788225528 love spells in Toronto Psychic Readings, Attraction spells,Brin...masabamasaba
 
Payment Gateway Testing Simplified_ A Step-by-Step Guide for Beginners.pdf
Payment Gateway Testing Simplified_ A Step-by-Step Guide for Beginners.pdfPayment Gateway Testing Simplified_ A Step-by-Step Guide for Beginners.pdf
Payment Gateway Testing Simplified_ A Step-by-Step Guide for Beginners.pdfkalichargn70th171
 
Direct Style Effect Systems - The Print[A] Example - A Comprehension Aid
Direct Style Effect Systems -The Print[A] Example- A Comprehension AidDirect Style Effect Systems -The Print[A] Example- A Comprehension Aid
Direct Style Effect Systems - The Print[A] Example - A Comprehension AidPhilip Schwarz
 
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfonteinmasabamasaba
 
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...SelfMade bd
 
%+27788225528 love spells in Atlanta Psychic Readings, Attraction spells,Brin...
%+27788225528 love spells in Atlanta Psychic Readings, Attraction spells,Brin...%+27788225528 love spells in Atlanta Psychic Readings, Attraction spells,Brin...
%+27788225528 love spells in Atlanta Psychic Readings, Attraction spells,Brin...masabamasaba
 
Architecture decision records - How not to get lost in the past
Architecture decision records - How not to get lost in the pastArchitecture decision records - How not to get lost in the past
Architecture decision records - How not to get lost in the pastPapp Krisztián
 
WSO2CON 2024 - WSO2's Digital Transformation Journey with Choreo: A Platforml...
WSO2CON 2024 - WSO2's Digital Transformation Journey with Choreo: A Platforml...WSO2CON 2024 - WSO2's Digital Transformation Journey with Choreo: A Platforml...
WSO2CON 2024 - WSO2's Digital Transformation Journey with Choreo: A Platforml...WSO2
 
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...WSO2
 
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...Steffen Staab
 
%in Rustenburg+277-882-255-28 abortion pills for sale in Rustenburg
%in Rustenburg+277-882-255-28 abortion pills for sale in Rustenburg%in Rustenburg+277-882-255-28 abortion pills for sale in Rustenburg
%in Rustenburg+277-882-255-28 abortion pills for sale in Rustenburgmasabamasaba
 

Dernier (20)

W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
 
Love witchcraft +27768521739 Binding love spell in Sandy Springs, GA |psychic...
Love witchcraft +27768521739 Binding love spell in Sandy Springs, GA |psychic...Love witchcraft +27768521739 Binding love spell in Sandy Springs, GA |psychic...
Love witchcraft +27768521739 Binding love spell in Sandy Springs, GA |psychic...
 
Large-scale Logging Made Easy: Meetup at Deutsche Bank 2024
Large-scale Logging Made Easy: Meetup at Deutsche Bank 2024Large-scale Logging Made Easy: Meetup at Deutsche Bank 2024
Large-scale Logging Made Easy: Meetup at Deutsche Bank 2024
 
%+27788225528 love spells in Knoxville Psychic Readings, Attraction spells,Br...
%+27788225528 love spells in Knoxville Psychic Readings, Attraction spells,Br...%+27788225528 love spells in Knoxville Psychic Readings, Attraction spells,Br...
%+27788225528 love spells in Knoxville Psychic Readings, Attraction spells,Br...
 
Abortion Pill Prices Tembisa [(+27832195400*)] 🏥 Women's Abortion Clinic in T...
Abortion Pill Prices Tembisa [(+27832195400*)] 🏥 Women's Abortion Clinic in T...Abortion Pill Prices Tembisa [(+27832195400*)] 🏥 Women's Abortion Clinic in T...
Abortion Pill Prices Tembisa [(+27832195400*)] 🏥 Women's Abortion Clinic in T...
 
What Goes Wrong with Language Definitions and How to Improve the Situation
What Goes Wrong with Language Definitions and How to Improve the SituationWhat Goes Wrong with Language Definitions and How to Improve the Situation
What Goes Wrong with Language Definitions and How to Improve the Situation
 
%+27788225528 love spells in Huntington Beach Psychic Readings, Attraction sp...
%+27788225528 love spells in Huntington Beach Psychic Readings, Attraction sp...%+27788225528 love spells in Huntington Beach Psychic Readings, Attraction sp...
%+27788225528 love spells in Huntington Beach Psychic Readings, Attraction sp...
 
%in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park %in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park
 
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
 
%+27788225528 love spells in Toronto Psychic Readings, Attraction spells,Brin...
%+27788225528 love spells in Toronto Psychic Readings, Attraction spells,Brin...%+27788225528 love spells in Toronto Psychic Readings, Attraction spells,Brin...
%+27788225528 love spells in Toronto Psychic Readings, Attraction spells,Brin...
 
Payment Gateway Testing Simplified_ A Step-by-Step Guide for Beginners.pdf
Payment Gateway Testing Simplified_ A Step-by-Step Guide for Beginners.pdfPayment Gateway Testing Simplified_ A Step-by-Step Guide for Beginners.pdf
Payment Gateway Testing Simplified_ A Step-by-Step Guide for Beginners.pdf
 
Direct Style Effect Systems - The Print[A] Example - A Comprehension Aid
Direct Style Effect Systems -The Print[A] Example- A Comprehension AidDirect Style Effect Systems -The Print[A] Example- A Comprehension Aid
Direct Style Effect Systems - The Print[A] Example - A Comprehension Aid
 
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
 
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...
Crypto Cloud Review - How To Earn Up To $500 Per DAY Of Bitcoin 100% On AutoP...
 
%+27788225528 love spells in Atlanta Psychic Readings, Attraction spells,Brin...
%+27788225528 love spells in Atlanta Psychic Readings, Attraction spells,Brin...%+27788225528 love spells in Atlanta Psychic Readings, Attraction spells,Brin...
%+27788225528 love spells in Atlanta Psychic Readings, Attraction spells,Brin...
 
Architecture decision records - How not to get lost in the past
Architecture decision records - How not to get lost in the pastArchitecture decision records - How not to get lost in the past
Architecture decision records - How not to get lost in the past
 
WSO2CON 2024 - WSO2's Digital Transformation Journey with Choreo: A Platforml...
WSO2CON 2024 - WSO2's Digital Transformation Journey with Choreo: A Platforml...WSO2CON 2024 - WSO2's Digital Transformation Journey with Choreo: A Platforml...
WSO2CON 2024 - WSO2's Digital Transformation Journey with Choreo: A Platforml...
 
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
 
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
 
%in Rustenburg+277-882-255-28 abortion pills for sale in Rustenburg
%in Rustenburg+277-882-255-28 abortion pills for sale in Rustenburg%in Rustenburg+277-882-255-28 abortion pills for sale in Rustenburg
%in Rustenburg+277-882-255-28 abortion pills for sale in Rustenburg
 

Diversifying Autocomplete

  • 1. 1 Diversifying Autocomplete Felipe Besson Haystack / MICES / Berlin Buzzwords June 11, 2020
  • 2. 2 20+ classifieds brands worldwide 350+ mi users each month 5000+ employees 35 offices Online Classifieds Platform 25+ countries
  • 3. OLX Europe Discovery Cycle 3 Recommendations home page search page ad page
  • 4. All steps are interconnected! 4 ● Users have different intents ● What can break the dialogue with the user ? ○ Broad queries (Autocomplete and Search) ○ Ambiguity (Query understanding) ○ Bad Interactions (Recommendations)
  • 5. Diversifying search results 5 Strength the dialogue with the user ● Dealing with broad queries ○ Autocomplete ○ Search ● Item showcase for new or exploring users ● Gathering more interactions to improve recommendations ● Autocomplete
  • 6. What will be covered and how ? 6 ● Broad queries problem in autocomplete ● Techniques to promote diversification ● Our use case: ○ Autocomplete at OLX Europe
  • 7. What is Autocomplete ? 7 A tool to talk directly to the user ● Guide users to good queries ● Help query understanding to understand ● Fast response/reaction ● Help tackling search relevance earlier as possible
  • 8. Autocomplete at OLX Europe 8 ● Suggest popular searches with category filters ● Covers 7 different countries ● > 50 mi requests per day ● Responsible for 40% of total searches ● Ranks suggestions by popularity and narrowness ○ but ...
  • 9. Broad query problem ... 9 What is my intent ? What if I don't know any Vespa model ? popularity What if I have a Vespa and want some accessory ?
  • 10. Broad query effect ... Fashion Bags and accessories Footwear Clothing Watches and Jewelry Notions Other bags and accessories Woman Sunglasses Man Woman Man Watches Jewelry 10 Different topics Level 1 (L1) Level 2 (L2) Level 3 (L3) Gucci Wallets Handbags Health and Beauty Perfumes Medical care Autocomplete suggestions
  • 11. Breaks in the dialogue with user 11 ● We jumped to premature conclusions ○ Show very specific popular suggestions (Vespa models) ● We could have asked more ○ Show more possibilities (like accessories) ● Maybe we will never have the chance to ask more ○ Popularity feedback loop ("rich get richer")
  • 12. Diversifying autocomplete suggestions 12 Improve user experience on broad queries ● Minimize overspecialization of suggestions ● Give an overview of different available item categories ● Break popularity feedback loop ● Refine the query (user intents)
  • 13. The goal 13 Diversifying autocomplete category suggestions for broad queries Broad queries = popular queries AND contain categories with many search results AND those categories are not yet suggested!
  • 14. How to apply diversification ? 14 Inspiration from Web Search and Information retrieval Explicit diversification ○ From query (information needs) ○ Increase Coverage ○ Broad queries Based on Search result diversification: http://www.dcs.gla.ac.uk/~craigm/publications/santos2015ftir.pdf
  • 15. How can we measure coverage ? 15 Step 1: Clustering documents into topics ○ Facets, categories, colors, word embeddings, ... 891 ... 36 ... 37 ... 903 3 topics topics probability
  • 16. How can we measure coverage ? 16 Step 2: Measure dispersion of topics distribution GINI Coefficient: https://opensourceconnections.com/blog/2019/09/05/diversity-vs-relevance <> GINI Coefficient Shannon Entropy topicstopics probability probability
  • 17. Shannon Entropy 17 Measures level of information in a probability distribution A B C High Knowledge Medium Knowledge Low Knowledge Low Surprise Medium Surprise High Surprise entropy = 0 entropy = 0.81 entropy = 1.5
  • 18. Shannon Entropy for e-commerces 18 1. Cluster document into categories (or any other criteria) 2. Category probability entropy: 2.38 entropy: 0.52
  • 19. Entropy from another perspective 19Extracted from: https://medium.com/udacity/shannon-entropy-information-gain-and-picking-balls-from-buckets-5810d35d54b4 On average, how many questions do we need to ask to find out what letter it is? Entropy = 0 Bucket 1 Entropy = 1.75 Bucket 2 Entropy = 2.0 Bucket 3 Akinator: https://en.wikipedia.org/wiki/Akinator
  • 20. Entropy from another perspective 20 Extracted from: https://medium.com/udacity/shannon-entropy-information-gain-and-picking-balls-from-buckets-5810d35d54b4 Bucket 3 (2 questions on overage) Bucket 2 (1.75 questions on average)
  • 21. Coming back to the autocomplete 21 On average, how many questions can we ask to make sure we cover all user intents ? each suggestion we give = a different question we make ○ 0 questions for very specific queries (low entropy) ○ n questions for broad queries ■ How many is n ? ■ How can we define these questions ?
  • 22. How many questions can we ask ? 22 possible question! entropy of each category 10 slots Entropy = # of different questions
  • 23. Maximum diversity is 10 different suggestions! ● Each category has p(x) = 0.1 and e(x) = 0.33 How to pick each suggestion ? 23 0.33 too few results Narrow queries candidates
  • 24. Generation new suggestions Fashion Bags and accessories Footwear Clothing Watches and Jewelry Notions 24 Gucci Health and Beauty H(X) = 1.27 p(x) = 0.56 e(x) = 0.47 p(x) = 0.15 e(x) = 0.41 p(x) = 0.14 e(x) = 0.39 p(x) = 0.09 e(x) = 0.32 p(x) = 0.05 e(x) = 0.22 p(x) = 0.002 e(x) = 0.019 L2
  • 25. Experiment pipeline 25 Goal: Expand suggestions for broad queries
  • 28. Experiment Scope 28 ● 2 countries (C1 and C2) ● Expansions for less than 5% of suggested queries but covered: ○ 26% of total searches for C1 ○ 17% of total searches for C2 ● Compared the performance of both groups ○ broad queries: expanded vs not expanded
  • 29. Primary metrics Description C1 C2 suggest_search_rate Autocomplete usage: # suggested searches / # total searches +10.41% +0.72% pos_filter_rate Search filters applied after picking expanded suggestions -3.14% -5.14% Experiment Results 29 ● Diversification impacted user behaviour in autocomplete ● C1 users interacted more with autocomplete suggestions ● Did C2 users pick less suggestions but better ones ?
  • 30. Experiment Results 30 Query metrics* Description C1 C2 suggest_ctr Uplift in ad clicks from expanded query +3.64% -3.86 suggest_reply_rate Uplift in ad replies from expanded query +1.81% +0.26% Suggestion metrics* Description C1 C2 suggest_cat_ctr Uplift in ad clicks from expanded suggestions (category) +2.24% +9.48% suggest_cat_reply_rate Uplift in ad replies from expanded suggestions (category) +6.13% +13.01% ● Promising for C1 users in general ● In C2, we might have replaced relevant suggestions ● In both countries, new suggested categories look relevant
  • 31. Considerations and Future 31 ● Early stage: first and simple iteration ● Extend experiment ○ Affect more queries and add more countries ● Impact short vs long term ○ Consider rank (top n results) ○ Explore more clustering dimensions ○ Define entropy and popularity thresholds (prior and observed)