Scale, Structure, and Semantics

This Semantic Technology & Business Conference (SemTechBiz) keynote argues that knowledge representation is overrated for AI systems and computation is underrated. It discusses past attempts at knowledge representation like Cyc and Freebase, and how today's data-driven approaches using large datasets have proven more effective than rule-based systems for tasks like machine translation and question answering. It advocates for semi-structured data and data-driven recommendations and queries to empower users and fill gaps in systems' knowledge. It concludes that communication is both the problem and solution, and systems should leverage users as intelligent partners rather than relying solely on perfect schemas or vocabularies.

Transcript

1. Recruiting Solutions Recruiting Solutions Recruiting Solutions Scale, Structure, and Semantics Daniel Tunkelang Principal Data Scientist at LinkedIn 1
2. Take-Aways 2 Communication trumps knowledge representation. Communication is the problem and the solution.
3. Overview 1. Knowledge representation is overrated. 1. Computation is underrated. 2. We have a communication problem. 3
4. The Bad News 1. Knowledge representation is overrated. 1. Computation is underrated. 2. We have a communication problem. 4
5. AI: a dream deferred. 5
6. Memex: the Computer Science Version 6
7. Cyc 7
8. Freebase 8
9. Wolfram Alpha 9
10. Knowledge representation is overrated. Today’s knowledge repositories are:  incomplete  inconsistent  inscrutable  and not sustained by economic incentives. 1986 estimate of effort to complete Cyc:  250,000 rules + 350 person-years 10
11. The Good News 1. Knowledge representation is overrated. 1. Computation is underrated. 2. We have a communication problem. 11
12. Deep Blue 12 vs.
13. Watson 13
14. Plain Old Search Engines are Pretty Good Too 14 http://blog.stephenwolfram.com/2011/01/jeopardy-ibm-and-wolframalpha/
15. The Unreasonable Effectiveness of Data  simple models + lots of data >> elaborate models + less data  machine translation: parallel corpora >> elaborate rules for syntactic and semantic patterns  semantic web formalism just means semantic interpretation on shorter strings between angle brackets Alon Halevy, Peter Norvig, and Fernando Pereira (2009) 15
16. Today’s Challenge 1. Knowledge representation is overrated. 1. Computation is underrated. 2. We have a communication problem. 16
17. Semi-structured Data 17 Michael K. Bergman, http://www.mkbergman.com/
18. Semi-structured Data at LinkedIn <person> <id> <first-name /> <last-name /> <location> <name> <country> <code> </country> </location> <industry> … </person> Summary I lead a data science team at LinkedIn, which analyzes terabytes of data to produce products and insights that serve LinkedIn’s members. Prior to LinkedIn, I led a local search quality team at Google and was a founding employee of faceted search pioneer Endeca (acquired by Oracle in 2010), where…
19. Semi-structured Search is a Killer App 19
20. Another Example: Helping a Friend Dear Daniel, I'm attaching the resume of an old friend who just moved up to the Bay Area. He has a very strong background in:  mobile / wireless applications  start-ups and new product launches  international expansion Best regards, XXX 20
21. Company Search 21
22. Semi-structured Data Empowers Users 22
23. Data-Driven Recommendations 23
24. Data-Driven Computation Serves Communication 24 for i in [1..n] s w1 w2 … wi if Pc (s) > 0 a new Segment() a.segs {s} a.prob Pc (s) B[i] {a} for j in [1..i-1] for b in B[j] s wj wj+1 … wi if Pc (s) > 0 a new Segment() a.segs b.segs U {s} a.prob b.prob * Pc (s) B[i] B[i] U {a} sort B[i] by prob truncate B[i] to size k
25. Recommendations Leverage Semi-structured Data 25 Corpus Stats Job 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 . . . 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 …
26. Skills: A Practical Knowledge Representation 26
27. Data-Driven Query Expansion for Recall 27
28. Data-Driven Query Refinement for Precision 28
29. There is no perfect schema or vocabulary.  And even if there were, not everyone would use it.  Knowledge representation has only succeeded within narrow scope.  Brute force is surprisingly effective but does not leverage the user as an intelligent partner. 29
30. Communication is the problem and the solution.  Rich communication channel fills gaps in system’s knowledge representation and in user’s knowledge.  Use data science to make the system smart, but be humble and empower the human user. You've got the brawn I've got the brains Let's make lots of money Pet Shop Boys, “Opportunities” 30
31. The Future is Upon Us 31
32. One More Thing “More data beats clever algorithms but better data beats more data.” Monica Rogati @ Strata 2012 32
33. Questions? Contact: dtunkelang@linkedin.com We’re Hiring! Thank You! 33