Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Large-Scale Social
Recommendation Systems:
Challenges And Opportunities
Mitul Tiwari!
!
Search, Network, and Analytics (SN...
Who Am I
2
Outline
• About LinkedIn!
• Social Recommender Systems at LinkedIn!
• Social Graph Analysis!
• Virality in Social Recommen...
Linkedin By The Numbers
4
225M members 2 new members/sec
Broad Range Of Products
5
Member Profile
6
Contacts
7
Talent Solutions
8
Job Search
9
Company Pages
10
Outline
• About LinkedIn!
• Social Recommender Systems at LinkedIn!
• Social Graph Analysis!
• Virality in Social Recommen...
Linkedin Homepage
• Powered by
recommendations
12
Recommender Ecosystem
13
!
Similar  Profiles
Connections
News
Skill  Endorsements
Outline
• Social Recommender Systems at LinkedIn!
• LinkedIn Today: Recommend News!
• Jobs Recommendation!
• Related Searc...
Linkedin Today: News Recommendation
• Objective: serve valuable professional news, leading to
higher engagement as measure...
News Recommendation: Explore/Exploit
16
item j from a set of candidates
User i
with
user features
(e.g., industry,
behavio...
News Recommendations: Challenges
• Drop in CTR wrt Time
17
News Recommendation: Challenges
• Same item shown to the same users: drop in CTR
18
News Recommendations: Revised Algorithm
• Explore/Exploit scheme!
• Explore: choose an item at random with a small probabi...
Outline
20
• Social Recommender Systems at LinkedIn!
• LinkedIn Today: Recommend News!
• Jobs Recommendations!
• Related S...
Jobs Recommendation
• Goal: recommend dream jobs to job
seekers!
• Challenges!
• Lag between view and application, offer,
...
Jobs Recommendation
22
17
Corpus StatsJob
User Base
Filtered
title
geo
company
industry
description
functional area
…
Cand...
Magic Is In Feature Engineering
• Open to relocation?!
• Region similarity based on profile or network!
• Region transition...
What Should You Transition To And When
24
• Probability of holding a title wrt time: spikes 12 months aligned
Wang et. al,...
Job Seeking: Socially Contagious
25
[Zhang, 2012]
• Prob. of quitting increases as the #of recently left connected colleag...
Outline
26
• Social Recommender Systems at LinkedIn!
• LinkedIn Today: Recommend News!
• Jobs Recommendation!
• Related Se...
Related Searches Recommendation
• Millions of Searches everyday!
• Help users to explore and refine their queries
27
Reda e...
Related Searches Recommendation
28
Related Searches Recommendations
• Signals!
• Collaborative Filtering!
• Query-Result Click graph!
• Overlapping terms!
• ...
Related Searches Recommendations
• Signals!
• Collaborative Filtering!
• Query-Result Click graph!
• Overlapping terms!
• ...
Outline
• About LinkedIn!
• Social Recommender Systems at LinkedIn!
• Social Graph Analysis!
• Virality in Social Recommen...
Link Prediction Over Social Graph
32
Inmaps: Connection Graph
33
Connection Strength
• Measure strength of each connection!
• Applications!
• Introductions!
• Update stream relevance
34
Query-Result Clicks Graph
• Application: Related Searches correlated by result clicks
for related searches recommendation
...
Skills Similarity Graph
• Graph of all co-occurrences between LinkedIn Skills
36
Skills
37
Find Influencers In Venture Capital?
38
Outline
• About LinkedIn!
• Social Recommender Systems at LinkedIn!
• Social Graph Analysis!
• Virality in Social Recommen...
Suggested Skill Endorsement
40
Skills Endorsements
41
Viral Growth: 1B Skills Endorsements
• One of the fastest growing product in LinkedIn’s history
42
Skill Tagging
• Tagging: extract potential skills from
profile using skills taxonomy!
!
!
• Standardize skill phrase varian...
Skill Recommendation
• Predict a skill even if not
present in the profile!
• Based on likelihood of
member having a skill!
...
Suggested Skill Endorsements
• Binary Classification!
• Features!
• Company overlap, School overlap,
Industrial and functio...
Social Recommendation And Tagging
SkillTagging
Skill Recommendation
Suggested Skill Endorsements
46
Skills Important For Data Scientists?
47
Outline
• About LinkedIn!
• Social Recommender Systems at LinkedIn!
• Social Graph Analysis!
• Virality in Social Recommen...
Scaling Challenges: Related Searches Example
• Kafka: publish-subscribe messaging system!
• Hadoop: MapReduce data process...
Outline
• About LinkedIn!
• Social Recommender Systems at LinkedIn!
• Social Graph Analysis!
• Virality in Social Recommen...
References
51
Acknowledgement
• Thanks to Data Team at LinkedIn: http://data.linkedin.com!
• We are hiring!!
• Contact: mtiwari[at]linke...
Questions?
53
Upcoming SlideShare
Loading in …5
×
Upcoming SlideShare
A Hybrid Recommendation system
Next

Share

Large-scale Social Recommendation Systems: Challenges and Opportunity

Keynote talk at 4th International Workshop on Social Recommender Systems (SRS 2013)
In conjunction with 22nd International World Wide Web Conference (WWW 2013). More details: http://cslinux0.comp.hkbu.edu.hk/~fwang/srs2013/

Related Books

Free with a 30 day trial from Scribd

See all

Related Audiobooks

Free with a 30 day trial from Scribd

See all

Large-scale Social Recommendation Systems: Challenges and Opportunity

  1. 1. Large-Scale Social Recommendation Systems: Challenges And Opportunities Mitul Tiwari! ! Search, Network, and Analytics (SNA)! LinkedIn
  2. 2. Who Am I 2
  3. 3. Outline • About LinkedIn! • Social Recommender Systems at LinkedIn! • Social Graph Analysis! • Virality in Social Recommender Systems! • Scaling Challenges 3
  4. 4. Linkedin By The Numbers 4 225M members 2 new members/sec
  5. 5. Broad Range Of Products 5
  6. 6. Member Profile 6
  7. 7. Contacts 7
  8. 8. Talent Solutions 8
  9. 9. Job Search 9
  10. 10. Company Pages 10
  11. 11. Outline • About LinkedIn! • Social Recommender Systems at LinkedIn! • Social Graph Analysis! • Virality in Social Recommender Systems! • Scaling Challenges 11
  12. 12. Linkedin Homepage • Powered by recommendations 12
  13. 13. Recommender Ecosystem 13 ! Similar  Profiles Connections News Skill  Endorsements
  14. 14. Outline • Social Recommender Systems at LinkedIn! • LinkedIn Today: Recommend News! • Jobs Recommendation! • Related Searches Recommendation! • Social Graph Analysis! • Virality in Social Recommender Systems! • Scaling Challenges 14
  15. 15. Linkedin Today: News Recommendation • Objective: serve valuable professional news, leading to higher engagement as measured by metrics such as CTR 15
  16. 16. News Recommendation: Explore/Exploit 16 item j from a set of candidates User i with user features (e.g., industry, behavioral features, Demographic features,……) (i, j) : response yijvisits Algorithm selects (click or not) Which item should we select? ! The item with highest predicted CTR ! An item for which we need data to predict its CTR Exploit Explore Agarwal et. al 2012
  17. 17. News Recommendations: Challenges • Drop in CTR wrt Time 17
  18. 18. News Recommendation: Challenges • Same item shown to the same users: drop in CTR 18
  19. 19. News Recommendations: Revised Algorithm • Explore/Exploit scheme! • Explore: choose an item at random with a small probability (e.g., 5%)! • Exploit: choose highest scoring CTR item (e.g., 95%)! • Temporal smoothing: more weight to recent data! • Impression discounting: discount items with repeat views! • Segmented model: segment users in CTR estimation! • Opportunity: Multi-arm bandit problem 19
  20. 20. Outline 20 • Social Recommender Systems at LinkedIn! • LinkedIn Today: Recommend News! • Jobs Recommendations! • Related Searches Recommendation! • Social Graph Analysis! • Social Update Stream and Virality! • Scaling Challenges
  21. 21. Jobs Recommendation • Goal: recommend dream jobs to job seekers! • Challenges! • Lag between view and application, offer, acceptance! • High level of expectations 21
  22. 22. Jobs Recommendation 22 17 Corpus StatsJob User Base Filtered title geo company industry description functional area … Candidate General expertise specialties education headline geo experience Current Position title summary tenure length industry functional area … Similarity (candidate expertise, job description) 0.56 Similarity (candidate specialties, job description) 0.2 Transition probability (candidate industry, job industry) 0.43 Title Similarity 0.8 Similarity (headline, title) 0.7 . . . derived Matching Binary Exact matches: geo, industry, … Soft transition probabilities, similarity, … Text Transition probabilities Connectivity yrs of experience to reach title education needed for this title … Ensemble Scorings Bhasin et. al 2012
  23. 23. Magic Is In Feature Engineering • Open to relocation?! • Region similarity based on profile or network! • Region transition probability! • Predict members’ propensity to migrate and potential regions 23
  24. 24. What Should You Transition To And When 24 • Probability of holding a title wrt time: spikes 12 months aligned Wang et. al, WWW’13
  25. 25. Job Seeking: Socially Contagious 25 [Zhang, 2012] • Prob. of quitting increases as the #of recently left connected colleague
  26. 26. Outline 26 • Social Recommender Systems at LinkedIn! • LinkedIn Today: Recommend News! • Jobs Recommendation! • Related Searches Recommendation! • Social Graph Analysis! • Social Update Stream and Virality! • Scaling Challenges
  27. 27. Related Searches Recommendation • Millions of Searches everyday! • Help users to explore and refine their queries 27 Reda et. al, CIKM’12
  28. 28. Related Searches Recommendation 28
  29. 29. Related Searches Recommendations • Signals! • Collaborative Filtering! • Query-Result Click graph! • Overlapping terms! • Length-bias! • Ensemble approach for unified recommendation! • Practical considerations 29
  30. 30. Related Searches Recommendations • Signals! • Collaborative Filtering! • Query-Result Click graph! • Overlapping terms! • Length-bias! • Ensemble approach for unified recommendation! • Practical considerations! • Opportunity: Build Personalized Search Recommendation 30
  31. 31. Outline • About LinkedIn! • Social Recommender Systems at LinkedIn! • Social Graph Analysis! • Virality in Social Recommender Systems! • Scaling Challenges 31
  32. 32. Link Prediction Over Social Graph 32
  33. 33. Inmaps: Connection Graph 33
  34. 34. Connection Strength • Measure strength of each connection! • Applications! • Introductions! • Update stream relevance 34
  35. 35. Query-Result Clicks Graph • Application: Related Searches correlated by result clicks for related searches recommendation Q1 Qn R1 Rm 35
  36. 36. Skills Similarity Graph • Graph of all co-occurrences between LinkedIn Skills 36
  37. 37. Skills 37
  38. 38. Find Influencers In Venture Capital? 38
  39. 39. Outline • About LinkedIn! • Social Recommender Systems at LinkedIn! • Social Graph Analysis! • Virality in Social Recommender Systems! • Scaling Challenges 39
  40. 40. Suggested Skill Endorsement 40
  41. 41. Skills Endorsements 41
  42. 42. Viral Growth: 1B Skills Endorsements • One of the fastest growing product in LinkedIn’s history 42
  43. 43. Skill Tagging • Tagging: extract potential skills from profile using skills taxonomy! ! ! • Standardize skill phrase variants Profile Tokenize SkillsTagger Phrases Skills 43
  44. 44. Skill Recommendation • Predict a skill even if not present in the profile! • Based on likelihood of member having a skill! • Features: company, industry, skills, ... 44 Profile Tokenize SkillsTagger Phrases Skills Skills Classifier Profile features Recommended Skills
  45. 45. Suggested Skill Endorsements • Binary Classification! • Features! • Company overlap, School overlap, Industrial and functional area similarity, Title similarity, Site interactions, Co- interactions, ... Candidate generation Classifier Features - Company - Title - Industry ... Suggested Endorsements 45
  46. 46. Social Recommendation And Tagging SkillTagging Skill Recommendation Suggested Skill Endorsements 46
  47. 47. Skills Important For Data Scientists? 47
  48. 48. Outline • About LinkedIn! • Social Recommender Systems at LinkedIn! • Social Graph Analysis! • Virality in Social Recommender Systems! • Scaling Challenges 48
  49. 49. Scaling Challenges: Related Searches Example • Kafka: publish-subscribe messaging system! • Hadoop: MapReduce data processing system ! • Azkaban: Hadoop workflow management tool! • Voldemort: Key-value store Metaphor Hadoop Search Backend Kafka Voldemort Related Searches Backend Front End HDFS 49
  50. 50. Outline • About LinkedIn! • Social Recommender Systems at LinkedIn! • Social Graph Analysis! • Virality in Social Recommender Systems! • Scaling Challenges 50
  51. 51. References 51
  52. 52. Acknowledgement • Thanks to Data Team at LinkedIn: http://data.linkedin.com! • We are hiring!! • Contact: mtiwari[at]linkedin.com! • Follow: @mitultiwari on Twitter 52 You! Applied Reseacher/ Research Engineer
  53. 53. Questions? 53
  • kennethowino9

    Jul. 15, 2018
  • yasminefathy

    Jul. 4, 2018
  • SungwookJeon1

    Apr. 29, 2016
  • vinay453

    Feb. 16, 2016
  • titleupz

    Nov. 13, 2015
  • chernbb

    Apr. 1, 2015
  • breezedeus

    Dec. 31, 2014

Keynote talk at 4th International Workshop on Social Recommender Systems (SRS 2013) In conjunction with 22nd International World Wide Web Conference (WWW 2013). More details: http://cslinux0.comp.hkbu.edu.hk/~fwang/srs2013/

Views

Total views

1,742

On Slideshare

0

From embeds

0

Number of embeds

57

Actions

Downloads

0

Shares

0

Comments

0

Likes

7

×