Big data innovation_summit_2014

40 %
60 %
Information about Big data innovation_summit_2014

Published on April 25, 2014

Author: AnujGoyal11



The talk I gave at the Big Data Innovation Summit.

Recruiting SolutionsRecruiting SolutionsRecruiting Solutions Recommending Jobs You May Be Interested In Anuj Goyal Recommendations at LinkedIn Anuj

§  About LinkedIn §  Search vs. Recommendations §  Recommendation Opportunities §  Evaluating Job Recommendations §  Job Recommendation Algorithm §  Challenges §  Summary 2 Overview

270+ M Company Pages >3M * Professional searches in 2012 ~5.7B 90%Fortune 100 Companies use LinkedIn to hire * *as of March 31, 2014 New Members joining ~2/sec 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 3 World’s largest professional network Over 65% of members are now international

4 Recommendation Versus Search Explicit Implicit

The Recommendations Opportunity 5


50%7 Value of Recommendations

Job Recommendations 8

9 Jobs You May be Interested In

Evaluation §  Upside metrics –  Are users getting relevant jobs? §  Downside metrics –  Are users getting offending jobs? 10

Evaluation - Upside §  Total Job Views (Clicks) §  Total Applications §  Total Viewers §  Total Applicants 11

Evaluation - Downside §  Applications per Click §  Clicks per Impression §  Applications per Impression §  Expert Judgments 12

User Features & Recommendation Algorithm

14 Positions Education Summary Experience Skills User Features

Corpus StatsCandidate Jobs User Base 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 Recommendation Algorithm Transition probabilities Connectivity yrs of experience to reach title education needed for this title … 15 Job Collection

Challenges & Feature Engineering

Challenge: Entity Resolution 17

‘IBM’ has 13000+ variations -  ibm – ireland -  ibm research -  T J Watson Labs -  International Bus. Machines Are All Companies The Same? 18

-  Software Engineer -  Technical Yahoo -  Member Technical Staff -  Software Development Engineer -  SDE Are All Titles The Same? 19

Same company with different name Same name but different companies Name Variations for IBM? “Orion” refers to 20 diff. companies large scale: 100M+ members, 2M+ company entities IBM: Intl Brotherhood of Magicians ~ 13000 Challenges – Entity Resolution 20

§  Binary classifier (LR), not ranker §  P({position, company entity} is a match) §  Features §  Content §  Social §  Behavior §  Company candidate set leveraged from Social graph and cosine similarity 97% Precision at 50% Coverage Asonam’11, KDD’11 Challenges – Entity Resolution 21 Precision Coverage

Challenges – Geo Location 22

§  Zip code mapped to Regions §  How sticky are those locations? Feature Engineering – Sticky locations 23

§  Open to relocation ? §  Region similarity based on profiles or network §  Region transition probability §  Predict individuals propensity to migrate and most likely migration target Feature Engineering – Sticky locations 24

Feature Engineering – The Network effect 25

Hybrid Recommendation Title : Research Engineer Company : Yahoo! Location : CA,USA Skills : Stats, ML, Java Title : Data Scientist Company : Samsung Location : PA,USA Skills : Stats, R Title : Analyst Company : Microsoft Location : CA, USA Skills : R, ML Title : Research Engineer <1>, Data Scientist <1>, Analyst <1> Company : Yahoo<1>, Samsung<1>, Microsoft<1> Location : CA,USA <2>, PA,USA<1> Skills : Stats<2>, ML<2>, R<2>, Java<1> Applicant Features Distribution Data Scientist / Senior Data Scientist San Jose 26

Information Gain Pick Top K overrepresented features from the applicants distribution A representative projection of the job in the member feature space 27 Hybrid Recommendation

§  Why Jobs Recommendations are Different §  Recommendation Algorithm §  Challenges –  Entity Resolution –  Location Resolution 28 Summary

Questions? Contact: We’re Hiring! Thank You! 29

Add a comment

Related presentations

How organisms adapt and survive in different environment.

Aplicación de ANOVA de una vía, modelo efectos fijos, en el problema de una empres...

Teori pemetaan

Teori pemetaan

November 10, 2014

learning how to mapping

Libros: Dra. Elisa Bertha Velázquez Rodríguez

Materi pelatihan gis

Materi pelatihan gis

November 10, 2014

learning GIS

In this talk we describe how the Fourth Paradigm for Data-Intensive Research is pr...

Related pages

Big Data Innovation Summit, Las Vegas

Big Data Technologies; Data Security; Data in the Cloud & much more... Request to Speak; View Schedule; More . Our Sponsors. Are you looking to showcase ...
Read more

Big Data Innovation Summit, London - The Innovation Enterprise

Big Data Innovation Summit #DataLondon16. Overview; Why Attend; Speakers; ... What is Big Data Innovation? Big Data Innovation has travelled, quite ...
Read more

Big Data Innovation Summit 2014: Highlights of Keynote ...

Highlights from keynote speeches by big data technology leaders from industry and academia on first day of Big Data Innovation Summit 2014 in Santa Clara.
Read more

Big Data Innovation Summit 2014: Highlights of Keynote ...

Highlights from keynote speeches by big data experts from Facebook, RedPoint Global, Quintiles, Samsung, GMU, PayPal, and others on Day 2 of Big Data ...
Read more

Big Data Innovation Summit 2014 | June 4th | clocate ...

Big Data Innovation Summit 2014 was held in Toronto, Canada on June 4th. Details on the event include dates, location and map, description, early ...
Read more

Big Data Innovation Summit 2014 - Clarity US

This Big Data Innovation Summit is sure to provide organized and competitive insight and solutions to Big Data challenges. Two days of Big Data information!
Read more

Day 2: Big Data Innovation Summit 2014 #DataWest14 | The ...

Hello again big data fans – from where I’ve learned the San Francisco 49’ers will be playing their 2014 NFL season at Levi’s Stadium… Santa Clara!
Read more

Big Data & Analytics Innovation Summit 2014 - Hong Kong

Guest Speakers: Opening Keynote Speaker: Panel Chairs: Mr. Alan Lau Senior Partner, Head of Technology Sector, Greater China McKinsey & Company: Mr ...
Read more