LinkedIn 2013 Annual Report Download - page 15

Download and view the complete annual report

Please find page 15 of the 2013 LinkedIn annual report below. You can navigate through the pages in the report by either clicking on the pages listed below, or by using the keyword search tool below to find specific information within the annual report.

Page out of 116

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80
  • 81
  • 82
  • 83
  • 84
  • 85
  • 86
  • 87
  • 88
  • 89
  • 90
  • 91
  • 92
  • 93
  • 94
  • 95
  • 96
  • 97
  • 98
  • 99
  • 100
  • 101
  • 102
  • 103
  • 104
  • 105
  • 106
  • 107
  • 108
  • 109
  • 110
  • 111
  • 112
  • 113
  • 114
  • 115
  • 116

Technology Infrastructure
Our technology platform is designed to create an engaging professional networking experience for our
members and is built to enable future growth at scale. We employ technological innovations whenever
possible to increase efficiency and scale our business.
Our products rely upon and leverage the massive amounts of data in our network. This rich dataset
has grown exponentially, requiring scalable computing resources. We will continue to invest in building
proprietary technologies and using open sourced technologies for our data, search and solutions. Our
product development expense was $395.6 million, $257.2 million and $132.2 million in 2013, 2012 and
2011, respectively.
Our key technology platforms are described below:
•Professional Graph. Our fully distributed system is comprised of a graph engine where nodes can
represent individuals, companies, schools and other entities and edges can be a connection, a
‘‘follow,’’ or an employee at a given company. The professional graph holds an individual’s
real-time network and enables a variety of complex calls like establishing the degree by which two
nodes are connected (e.g., 2nd degree vs. 3rd degree).
•Open Sourced Technologies. We deploy aspects of our technology into the open source community to
help increase the speed at which the technology can mature. The combination of open source and
proprietary technologies used in our platforms increases the speed at which we can deploy our
products at scale. For example, Hadoop is an open source project used to batch compute data for
different features on our website based on our members’ data and traffic patterns. Hadoop enables
us to scale our calculations on an expanding set of data and to perform these calculations more
frequently.
•Search. Our proprietary search technology combines structured and free-form content to allow
users to search across numerous parameters. Our search is powered by our rich dataset based on
facets and keywords and is fundamentally personalized as all search requests use a member’s
network to affect relevance and ranking. Our search is real time, distributed and multilingual and
serves the needs of both members and enterprises and professional organizations.
•Customized Content, Matching, Targeting and Recommendations. We have developed a proprietary
intelligence and recommendation engine for extracting professional insights by utilizing our rich
dataset. This engine enables us to provide our users with customized content and
recommendations. For example, based on a member’s profile, their second and third degree
connections, their viewing and clicking history, and a host of other criteria, our algorithms can
provide intelligence and recommendations around Talent Match, People You May Know, Groups
You May Like, Jobs You May Be Interested In, Sponsored Content or Companies You May Be
Interested In. Our targeting and recommendation technologies continue beyond just a member’s
profile by providing intelligence around similar profile views and similar job views.
•Ad Targeting Platform. We use a combination of traditional and proprietary ad targeting and
delivery technologies. The combination is optimized to work with our respective partners to provide
the optimal user experience. Our proprietary systems leverage our feature extraction, information
retrieval, and matching systems to provide the most relevant ads.
•Document Conversion Technologies. We use a combination of open source and proprietary
technologies to convert documents in various formats (e.g., pdf, doc, ppt) into HTML5, a mark-up
language for structuring and presenting content on the Internet, so that the document can be
displayed on LinkedIn.com and Slideshare.net, and embedded throughout the Internet.
13