![Job Seeking: Socially Contagious
25
[Zhang, 2012]
? Prob. of quitting increases as the #of recently left connected colleague](https://image.slidesharecdn.com/srskeynote-140817220631-phpapp01/85/Large-scale-Social-Recommendation-Systems-Challenges-and-Opportunity-25-320.jpg)
![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](https://image.slidesharecdn.com/srskeynote-140817220631-phpapp01/85/Large-scale-Social-Recommendation-Systems-Challenges-and-Opportunity-52-320.jpg)
Large-scale Social Recommendation Systems: Challenges and Opportunity
- 1. Large-Scale Social Recommendation Systems: Challenges And Opportunities Mitul Tiwari! ! Search, Network, and Analytics (SNA)! LinkedIn
- 2. Who Am I 2
- 3. Outline ? About LinkedIn! ? Social Recommender Systems at LinkedIn! ? Social Graph Analysis! ? Virality in Social Recommender Systems! ? Scaling Challenges 3
- 4. Linkedin By The Numbers 4 225M members 2 new members/sec
- 5. Broad Range Of Products 5
- 7. Contacts 7
- 9. Job Search 9
- 10. Company Pages 10
- 11. Outline ? About LinkedIn! ? Social Recommender Systems at LinkedIn! ? Social Graph Analysis! ? Virality in Social Recommender Systems! ? Scaling Challenges 11
- 12. Linkedin Homepage ? Powered by recommendations 12
- 13. Recommender Ecosystem 13 ! Similar ?Pro?les Connections News Skill ?Endorsements
- 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. Linkedin Today: News Recommendation ? Objective: serve valuable professional news, leading to higher engagement as measured by metrics such as CTR 15
- 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. News Recommendations: Challenges ? Drop in CTR wrt Time 17
- 18. News Recommendation: Challenges ? Same item shown to the same users: drop in CTR 18
- 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. 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. Jobs Recommendation ? Goal: recommend dream jobs to job seekers! ? Challenges! ? Lag between view and application, offer, acceptance! ? High level of expectations 21
- 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. Magic Is In Feature Engineering ? Open to relocation?! ? Region similarity based on pro?le or network! ? Region transition probability! ? Predict members’ propensity to migrate and potential regions 23
- 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. Job Seeking: Socially Contagious 25 [Zhang, 2012] ? Prob. of quitting increases as the #of recently left connected colleague
- 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. Related Searches Recommendation ? Millions of Searches everyday! ? Help users to explore and re?ne their queries 27 Reda et. al, CIKM’12
- 29. Related Searches Recommendations ? Signals! ? Collaborative Filtering! ? Query-Result Click graph! ? Overlapping terms! ? Length-bias! ? Ensemble approach for uni?ed recommendation! ? Practical considerations 29
- 30. Related Searches Recommendations ? Signals! ? Collaborative Filtering! ? Query-Result Click graph! ? Overlapping terms! ? Length-bias! ? Ensemble approach for uni?ed recommendation! ? Practical considerations! ? Opportunity: Build Personalized Search Recommendation 30
- 31. Outline ? About LinkedIn! ? Social Recommender Systems at LinkedIn! ? Social Graph Analysis! ? Virality in Social Recommender Systems! ? Scaling Challenges 31
- 32. Link Prediction Over Social Graph 32
- 34. Connection Strength ? Measure strength of each connection! ? Applications! ? Introductions! ? Update stream relevance 34
- 35. Query-Result Clicks Graph ? Application: Related Searches correlated by result clicks for related searches recommendation Q1 Qn R1 Rm 35
- 36. Skills Similarity Graph ? Graph of all co-occurrences between LinkedIn Skills 36
- 37. Skills 37
- 38. Find In?uencers In Venture Capital? 38
- 39. Outline ? About LinkedIn! ? Social Recommender Systems at LinkedIn! ? Social Graph Analysis! ? Virality in Social Recommender Systems! ? Scaling Challenges 39
- 42. Viral Growth: 1B Skills Endorsements ? One of the fastest growing product in LinkedIn’s history 42
- 43. Skill Tagging ? Tagging: extract potential skills from pro?le using skills taxonomy! ! ! ? Standardize skill phrase variants Pro?le Tokenize SkillsTagger Phrases Skills 43
- 44. Skill Recommendation ? Predict a skill even if not present in the pro?le! ? Based on likelihood of member having a skill! ? Features: company, industry, skills, ... 44 Pro?le Tokenize SkillsTagger Phrases Skills Skills Classi?er Pro?le features Recommended Skills
- 45. Suggested Skill Endorsements ? Binary Classi?cation! ? Features! ? Company overlap, School overlap, Industrial and functional area similarity, Title similarity, Site interactions, Co- interactions, ... Candidate generation Classi?er Features - Company - Title - Industry ... Suggested Endorsements 45
- 46. Social Recommendation And Tagging SkillTagging Skill Recommendation Suggested Skill Endorsements 46
- 47. Skills Important For Data Scientists? 47
- 48. Outline ? About LinkedIn! ? Social Recommender Systems at LinkedIn! ? Social Graph Analysis! ? Virality in Social Recommender Systems! ? Scaling Challenges 48
- 49. Scaling Challenges: Related Searches Example ? Kafka: publish-subscribe messaging system! ? Hadoop: MapReduce data processing system ! ? Azkaban: Hadoop work?ow management tool! ? Voldemort: Key-value store Metaphor Hadoop Search Backend Kafka Voldemort Related Searches Backend Front End HDFS 49
- 50. Outline ? About LinkedIn! ? Social Recommender Systems at LinkedIn! ? Social Graph Analysis! ? Virality in Social Recommender Systems! ? Scaling Challenges 50
- 51. References 51
- 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. Questions? 53