Dealing with structured and unstructured data at Facebook thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Dealing with structured and unstructured data at Facebook

Published on Jul 07, 201114138 Views

Facebook has undergone tremendous growth in the last five years. Here we will start by looking at some basic statistics and trends that have accompanied this growth. We'll then dive into two different

Related categories

Chapter list

facebook00:00
Dealing with Structured and Unstructured Data at Facebook01:05
Agenda - 101:16
Agenda - 201:50
500 million 30-day active users01:52
Over 30 billion pieces of content shared every month02:26
Over 3 billion photos uploaded each month02:58
1 million websites using Facebook platform03:15
Generated Data03:39
Profiles (ca. 2008)04:22
Profiles (2011)05:02
Structured and Unstructured Data06:06
The Friendship graph08:28
Pages08:58
Photos09:32
Places09:53
Open Graph Pages10:40
More than just nodes and edges11:14
Challenges - 112:04
Challenges - 214:07
Why Bother?15:58
Data Takeaways17:10
Agenda - 318:04
What are users talking about? - 118:06
What are users talking about? - 219:28
FML?19:56
Enriched Status Updates20:49
Unstructured Updates + Structured Data22:10
Vodka Map23:10
Page Like Graph (O'Connor)24:23
Finding Related Pages - 126:10
Finding Related Pages - 227:21
"Israel" Page - Page graph - 127:47
"Israel" Page - Page graph - 228:34
Ranking - 129:13
Ranking - 231:04
Feedback vs. Distance32:40
Checkin Heatmaps34:22
San Francisco Heatmap34:50
San Francisco Heatmap (Male)35:10
San Francisco Heatmap (Female)35:36
San Francisco Heatmap (Republican)35:59
San Francisco Heatmap (Democrat)36:20
Agenda - 436:31
People you may know37:17
Helping people find friends on FB38:03
How to make suggestions - 141:31
How to make suggestions - 242:30
How to make suggestions - 343:06
Suggesting Friends of Friends43:22
Friends in Common44:18
System Overview47:20
Agenda - 549:03
Making Static Predictions49:04
Friend of Friend Features49:44
Showing the best suggestion every time50:40
Reranking with logistic regression - 151:23
Reranking with logistic regression - 251:24
Putting it all together51:25
Agenda - 651:28
Performance51:30
Summary52:48
Questions54:13