22期.百度彭滔 搜索引擎评估与用户行为分析

755 views

Published on

baidu search engine user experience analyst

Published in: Technology, News & Politics
0 Comments
3 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
755
On SlideShare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
20
Comments
0
Likes
3
Embeds 0
No embeds

No notes for slide

22期.百度彭滔 搜索引擎评估与用户行为分析

  1. 1. 2012-01-07
  2. 2. Who am I!  Who am I –  –  pengtao@baidu.com –  •  –  •  •  "
  3. 3. 1.!  –  2010 81.9% •  CNNIC, 2011 –  Google effects on memory •  v.s. •  v.s. –  (Sparrow, 2011) •  The Internet has become a primary form of external or transactive memory, where information is stored collectively outside ourselves.
  4. 4. 1.!  –  ​1⁄1 × 2 •  Query #$ url + –  ​1⁄2 × 3 + •  ​1⁄3 × 1!  + –  MAP ​1⁄4 × 2 + –  DCG ​1⁄5 × 2 + –  nDCG –  ERR ​1⁄6 × 2 = –  … 5.0667
  5. 5. 2.!  Side by side
  6. 6. 2.!  –  •  E v.s. C –  10000 •  log 10000 query E C –  1000 •  10000 query 1000 –  100 •  1000 diff PM 100 review •  30 good) : 50 (same) : 20 (bad) PM
  7. 7. 2.!  –  •  v.s. pm query –  •  “ ” PM
  8. 8. 3.!  crowdsourcing) –  evaluator) –  –  evaluator!  WSE –  –  –  – 
  9. 9. 2.!  WSE evaluator
  10. 10. 3.!  Lesson1: – 
  11. 11. 3.!  Lesson2: –  – 
  12. 12. 3.!  Lesson3: –  –  •  Economics –  •  •  evaluator
  13. 13. 3.!  WSE –  –  10w!  crowdsourcing –  reCaptcha –  Amazon Mechanical Turk –  ESP Game –  Human computation
  14. 14. 3.!  –  –  AB testing, Bucket testing 50% % 50% 100%
  15. 15. 3.!  AB testing ? – 
  16. 16. 3.!  AB testing ? – 
  17. 17. 3.!  AB testing –  + –  –  – 
  18. 18. 3.!  AB testing –  1T •  cubeproducer, disql hadoop –  1G olap •  infobright, mondrian –  1M •  ABreport
  19. 19. 3.!  AB testing –  •  –  Overall Evaluation Criteria »  ("Crook,"2009)" –  Queryrank: »  ) »  ) •  –  – 
  20. 20. 3.!  –  •  –  •  –  AA test –  •  –  – 
  21. 21. 3.!  – 
  22. 22. 3.!  –  50% v.s. 50% –  B1 a1 i1 u1 Baidu i2 d1 B2 i3 u2 a2 B3 d2 i4 B4 d3 u3
  23. 23. !  –  –  DCG!  –  PM review –  crowdsourcing –  AB testing
  24. 24. !  –  v.s. AB testing –  –  v.s.
  25. 25. 1 wse
  26. 26. 2 ) ) sid X X’/Cookie Sid=1001) M1 BWS M2 N2 User)log internal)log M10 BWS
  27. 27. 关注我们:t.baidu-tech.com 资料下载和详细介绍:infoq.com/cn/zones/baidu-salon“畅想•交流•争鸣•聚会”是百度技术沙龙的宗旨。 百度技术沙龙是由百度与InfoQ中文站定期组织的线下技术交流活动。目的是让中高端技术人员有一个相对自由的思想交流和交友沟通的的平台。主要分讲师分享和OpenSpace两个关键环节,每期只关注一个焦点话题。讲师分享和现场Q&A让大家了解百度和其他知名网站技术支持的先进实践经验,OpenSpace环节是百度技术沙龙主题的升华和展开,提供一个自由交流的平台。针对当期主题,参与者人人都可以发起话题,展开讨论。 InfoQ 策划·组织·实施 关注我们:weibo.com/infoqchina

×