Ca

914 views

Published on

Published in: Technology, Business
0 Comments
5 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
914
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
13
Comments
0
Likes
5
Embeds 0
No embeds

No notes for slide

Ca

  1. 1.
  2. 2.
  3. 3.
  4. 4. agenda<br />Content match basics<br />Ad selection<br />Ad selection basics<br />Finding advertising keywords on web pages<br />Holistic view at the page in Contextual Advertising<br />Using click data to improve IR ad retrieval<br />When to advertise (relevance threshold)<br />the complete pipeline<br />
  5. 5.
  6. 6.
  7. 7. 谷歌<br />2003年3月, google发布adsense<br />2004年10月,进入中国<br />AdSense for Feeds<br />AdSense for search<br />AdSense for mobile content<br />AdSense for domains<br />AdSense for video<br />雅虎<br />2003年6月,Overture公司发布CM广告(Content Match); 雅虎YPN(Yahoo Publisher Network)<br />2007年, 推出新一代CM平台keystone<br />微软<br />2006年,微软推出广告计划adcenter<br />百度<br />2005年,百度推出主题推广CPRO<br />窄告, 2004年推出<br />阿里妈妈,2007年推出<br />
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18. agenda<br />Content match basics<br />Ad selection<br />Ad selection basics<br />Finding advertising keywords on web pages<br />Holistic view at the page in Contextual Advertising<br />Using click data to improve IR ad retrieval<br />When to advertise (relevance threshold)<br />the complete pipeline<br />
  19. 19. Ad selection basics<br />Objective<br />Database approach and IR methods<br />Understanding the pages, understanding the ads, matching<br />A few considerations<br />
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40. agenda<br />Content match basics<br />Ad selection<br />Ad selection basics<br />Finding advertising keywords on web pages<br />Holistic view at the page in Contextual Advertising<br />Using click data to improve IR ad retrieval<br />When to advertise (relevance threshold)<br />the complete pipeline<br />
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52. agenda<br />Content match basics<br />Ad selection<br />Ad selection basics<br />Finding advertising keywords on web pages<br />Holistic view at the page in Contextual Advertising<br />Using click data to improve IR ad retrieval<br />When to advertise (relevance threshold)<br />the complete pipeline<br />
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.
  67. 67.
  68. 68. agenda<br />Content match basics<br />Ad selection<br />Ad selection basics<br />Finding advertising keywords on web pages<br />Holistic view at the page in Contextual Advertising<br />Using click data to improve IR ad retrieval<br />When to advertise (relevance threshold)<br />the complete pipeline<br />
  69. 69. Using click data to improve IR ad retrieval<br />
  70. 70. Why CTR model/prediction<br />Traditional IR is not enough<br />Foundation for ad serving optimization<br />Building block for<br />CPM calculation<br />Quality score<br />
  71. 71.
  72. 72. CTR Modeling/Prediction<br />Will the user click on the ad at exactly this time at this page?<br />CTR(u, a, p) ,click-through rate, very small <br />Separate into three parts:<br />Feature selection (first extract, then select)<br />Model building<br />Evaluation metrics<br />Focusing on features<br />Scalable model building is key<br />
  73. 73. remarks<br />
  74. 74. agenda<br />Content match basics<br />Ad selection<br />Ad selection basics<br />Finding advertising keywords on web pages<br />Holistic view at the page in Contextual Advertising<br />Using click data to improve IR ad retrieval<br />When to advertise (relevance threshold)<br />the complete pipeline<br />
  75. 75.
  76. 76.
  77. 77.
  78. 78.
  79. 79.
  80. 80. agenda<br />Content match basics<br />Ad selection<br />Ad selection basics<br />Finding advertising keywords on web pages<br />Holistic view at the page in Contextual Advertising<br />Using click data to improve IR ad retrieval<br />When to advertise (relevance threshold)<br />the complete pipeline<br />
  81. 81. The complete pipeline<br />Ad selection and pricing<br />Serving pipe and building pipe<br />Request understanding, ads understanding, matching<br />Trigger, rank1, rank2, rank3…<br />……<br />
  82. 82. The funnel model<br />
  83. 83. Thank you<br />dingcanbiao@gmail.com<br />www.douban.com/people/canbiao/<br />

×