Kelvin Newman - Graph Theory

Transcript

1. LINK GRAPH SOCIAL GRAPH KNOWLEDGE GRAPH ETC. graph theory ! The theory that underpins how all search engines work @kelvinnewman
2. Graph Theory The most important theory in search that nobody talks about Kelvin Newman @kelvinnewman JD Hancock
3. Organiser BrightonSEO / Content Marketing Show / MeasureFest Three Free (and awesome) Conferences
4. Strategy Director SiteVisibility A digital agency specialising in retail, travel and financial services
5. Co-Founder Clockwork Talent Decent Digital Recruitment
6. shhhhh! JD Hancock
7. I might get in trouble for this JD Hancock JD Hancock
8. I’ve been let into a secret future beta of Google, and I’m going to reveal it to you
10. Joking aside; I think FB GraphSearch is a great indicator of the future of G JD Hancock
11. as it helps us better understand one of the theories that underlies all search engines JD Hancock
12. graph theory !
13. graph theory ! Hugely Important
14. graph theory ! Rarely Spoken About
15. LINK GRAPH
16. LINK GRAPH SOCIAL GRAPH
17. LINK GRAPH SOCIAL GRAPH KNOWLEDGE GRAPH
18. LINK GRAPH SOCIAL GRAPH KNOWLEDGE GRAPH OPEN GRAPH
19. LINK GRAPH SOCIAL GRAPH KNOWLEDGE GRAPH SORT OF
20. LINK GRAPH SOCIAL GRAPH KNOWLEDGE GRAPH ETC.
21. LINK GRAPH SOCIAL GRAPH KNOWLEDGE GRAPH ETC. graph theory ! Hugely Important
22. but we’ve been distracted, from what our job really is
23. which is understanding how the search engines fundamentally work jronaldlee
24. LINK GRAPH SOCIAL GRAPH KNOWLEDGE GRAPH ETC. graph theory ! is central to that understanding
25. this presentation may contain maths Benson Kua
26. hartjeff12
27. will - cambridge maths Dana Lookadoo - Yo! Yo! SEO Will Critchlow Maths MA - University of Cambridge
28. Tom Anthony PhD Artificial Intelligence University of Hertfordshire
29. Kelvin Newman Media Studies University of Sussex
30. I’m no computer scientist or mathematician jlwo
31. “a mathematical model for any system involving a binary relation” graph theory ! Frank Harary, 1969
32. “perhaps even more than to the contact between mankind and nature, graph theory owes to the contact of human beings between each other” Dénes König, 1936
33. http://www.slideshare.net/digitalmethods/gephi-rieder-23834788
34. Vertices or Nodes dominicotine
35. edges Matt Seppings
36. Nodes are Nouns
37. Edges are Verbs
38. basic graph visualisation
42. This Graph is Isomorphic of the other one, aka it’s the same but looks different
43. Matrix view of graph V1 V2 V3 V4 V5 V6 V1 V2 V3 V4 V5 V6
44. Matrix view of graph V1 V2 V3 V4 V5 V6 V1 0 V2 V3 V4 V5 V6
45. Matrix view of graph V1 V2 V3 V4 V5 V6 V1 0 1 V2 V3 V4 V5 V6
46. Matrix view of graph V1 V2 V3 V4 V5 V6 V1 0 1 1 1 V2 V3 V4 V5 V6
47. Matrix view of graph V1 V2 V3 V4 V5 V6 V1 0 1 1 1 0 0 V2 V3 V4 V5 V6
48. Matrix view of graph V1 V2 V3 V4 V5 V6 V1 0 1 1 1 0 0 V2 1 0 V3 V4 V5 V6
49. Matrix view of graph V1 V2 V3 V4 V5 V6 V1 0 1 1 1 0 0 V2 1 0 0 0 0 0 V3 1 0 0 0 0 0 V4 V5 V6
50. Matrix view of graph V1 V2 V3 V4 V5 V6 V1 0 1 1 1 0 0 V2 1 0 0 0 0 0 V3 1 0 0 0 0 0 V4 1 0 0 0 1 0 V5 0 0 0 1 0 1 V6 0 0 0 0 1 0
51. Or maybe? Blogger 1 Blogger 2 Blogger 3 Blogger 4 Blogger 5 Blogger 6 Blogger1 0 1 1 1 0 0 Blogger 2 1 0 0 0 0 0 Blogger 3 1 0 0 0 0 0 Blogger 4 1 0 0 0 1 0 Blogger 5 0 0 0 1 0 1 Blogger 6 0 0 0 0 1 0
52. Cardinality is the number of Nodes or Vertices in a Graph Prayitno/
53. Degrees of Vertex is how many edges a vertex has. chedder
54. Trees & Circuits Our Graph here is known as a tree, because you can’t loop back on yourself. If you could loop back on yourself it would be known as a circuit This is interesting to think about in the context of your site, or an area of the link graph
55. Watch PatrickJMT’s Graph Theory Videos http://patrickjmt.com/graph-theory-an-introduction/
56. What’s the first thing you teach your team? http://i.imgur.com/PGE2D2n.gif
57. http://computationalculture.net/article/what_is_in_pagerank For me it is PageRank
58. http://computationalculture.net/article/what_is_in_pagerank Jim Seward is a legend
59. What is PageRank? http://i.imgur.com/aNXqGNT.gif
60. A set of rules which can be used to give a numerical weighting to assess the importance of document within linked data set
61. A set of rules which can be used to give a numerical weighting to assess the importance of document within linked data set nodes
62. it is not
63. PageRank is used for than the Algo natalielucier
64. Understand Lung Cancer jasleen_kaur http://www.news-medical.net/news/20130326/Algorithm-similar-to-Google-PageRank-helps-map-spread-of-lung-cancer.aspx
65. Rank Scientific Significance http://bulib4research.blogspot.co.uk/2008/11/eigenfactor-scimago-journal-rankings.html
66. Predict Traffic deepsan http://iopscience.iop.org/1742-5468/2008/07/P07008/
67. three different surfers Chris Hunkeler
68. three different surfers Chris Hunkeler Random Surfer
69. Random Surfer Reflects the chance that the random surfer will leave the site through a link chosen at random, so all equally likely, and therefore valuable
70. three different surfers Chris Hunkeler Reasonable Surfer
71. Reasonable Surfer The reasonable surfer model supposes that some links are more likely to be clicked on and therefore should be given more value.
72. three different surfers Chris Hunkeler Intentional Surfer
73. Intentional Surfer The intentional surfer model supposes that links which ‘actually’ receive the most links should be given more value. http://en.wikipedia.org/wiki/PageRank#The_intentional_surfer_model
74. A lot has changed at Google, but it will always be a search engine which relies upon PageRank; which is a practical application of Graph Theory
75. Insert Audience Participation
76. Hands up who thinks FB GraphSearch is the best search engine in the world?
77. Just me?
78. Not here to convince you GraphSearch will catch on but...
79. If the area of this slide represents all the traffic on the internet
80. This much is Facebook http://mashable.com/2010/11/19/facebook-traffic-stats/
81. And every thing in white is the rest of the internet
82. Google, YouTube, Wikipedia, The Daily Mail, etc.
83. your website, my website, her website etc.
84. If anyone can build a Google-Killer it’s Facebook...
85. There’s a fundamental difference between Facebook & Google
86. is about...
87. documents and links JD Hancock
88. is about...
89. JD Hancock things and relationships
90. this difference is subtle but huge
91. but I think it works better for the web as we know it JD Hancock
92. Google are trying to catch- up but will struggle zoom images
93. JD Hancock Facebook’s data has a far more explicit structure than traditional web text
94. it’s not that tricky for Google to parse “I Like Nerf Guns” porkist
95. they could even have a go at “I was at Bodeans on Poland Street for Lunch Yesterday”* *if you mark it up in the right way R_Savvy
96. but has a much harder job understanding “Kelvin is married to Carolyn”
97. Facebook knows that happened in 2007
98. And who attended the ceremony
99. And when we got engaged
100. etc.
101. Google have to infer structure
102. Facebook know the structure
103. On GraphSearch you’re not really making a search. You’re just filtering a structured database of all the data Facebook has.
104. On GraphSearch you’re not really making a search. You’re just filtering a structured database of all the data Facebook has.
105. But it’s a bloody big database JD Hancock
106. 1 Billion Users Every Month
107. 240 Million Photo’s Per Day
108. 2.7 Billion Likes Everyday
109. People share billions of pieces of content everyday
110. One trillion connections of a thousand different types
111. 1,000,000,000,000
112. http://mashable.com/2013/07/08/facebook-launch-graph-search/
113. Every User, Page, Photo, Post & Place is a Node https://thetribe.s3.amazonaws.com/ferris.gif
114. Every friendship, checkin, tag or like is an Edge http://maxlutz.com/blog/wp-content/uploads/2013/05/coffee2.gif
115. Each Node has Meta-Data like description, this how the old FB Search “worked”
117. GraphSearch Allows you search the Edges as well as the Nodes JD Hancock JD Hancock
118. GraphSearch makes it easy to find nodes that are connected to another node by searching for an edge-type combined with an input node. E.g.: ▪Your friends: friend:10003 ▪People who live in new york: lives-in:111 ▪People who like downtown abbey: like:222
119. ‘Facebook use query-independent signals to come up with a numeric value for importance. This value is called the “static rank” of the entity.’ JD Hancock
120. What makes up static rank is still up for debate, but sensibly could be informed by the elements of Edgerank aka the (old name for) newsfeed algo
122. Affinity
123. Weight
124. Decay
125. The value of legitimate likes from well connected people just increased
126. There’s also been lot going on at Google
127. not a new update Martin Cathrae
128. but a new paradigm
129. introducing Knowledge Graph*
130. *and things not technically Knowledge Graph but sort of along the same lines introducing Knowledge Graph*
131. The Knowledge Graph enables you to search for things, people or places that Google knows about—landmarks, celebrities, cities, sports teams, buildings, geographical features, movies, celestial objects, works of art and more—and instantly get information that’s relevant to your query. Amit Singhal, Google
132. Knowledge Graph is part of a huge change in how Google deliver search results
133. I’m now going to give you lots of examples of changes in the way Google present results, not all of them are truly ‘Knowledge Graph’ but do indicate a general shift in the way they present results.
138. There’s more than 85 of these features that Dr. Pete from Moz has documented http://www.slideshare.net/crumplezone/ beyond-10-blue-links-the-future-of-ranking
139. But they’re just for informational queries... Right?
146. a change in purpose: help find pages help find answers
147. a change in purpose: help find pages help find answers
148. You can no longer rely on Google to send you traffic
149. or even tell you about it Alex E. Proimos
150. JD Hancock for nearly a year iPhone search traffic appeared as direct
151. and we’re rapidly approaching the point where we have no data on keyword traffic
152. Search isn’t about keywords anymore
153. It's about entities. chukgawlikphotography
154. Entities are normally, people, places, brands etc JD Hancock
155. but can be any ‘thing’ which has a relationship to another ‘thing’ JD Hancock
156. how can you make money if nobody ever goes to your site? JD Hancock
157. You may need to revisit your business model kennymatic
158. I love the Business Model Canvas http://en.wikipedia.org/wiki/Business_Model_Canvas
159. sit down and ask yourself could your business have an api
160. as every business is really just a database and a front end JD Hancock
161. and Google wants to become that front-end JD Hancock
162. So what can I do?
163. Familiarize yourself with Freebase http://www.freebase.com/
164. And DBpedia http://wiki.dbpedia.org/Datasets
165. It’s amazing the data they have yaph
166. If any of your keywords contain entities you MUST be prepared http://i.imgur.com/GLCC0bd.gif
167. Use BlueNod to Visualise Social Networks http://bluenod.com/
168. Different communities manifest themselves in different ways http://www.beautifullife.info/wp-content/uploads/2012/12/11/05.gif
172. Play with VisualDataWeb http://www.visualdataweb.org/relfinder
174. Mark Up using the Open Graph Protocol http://ogp.me/
175. Implement Schema.org http://schema.org/
176. No schema? Create one/extend one http://schema.org/docs/extension.html
177. Follow Peter Mika @pmika
178. Read Matthew J. Brown’s Mozcon Deck http://www.slideshare.net/MatthewBrownPDX/strings-to-things-the-move-to-semantic-seo-mozcon-2013
179. Watch WSDM Videos Web Search and Data Mining Conference http://videolectures.net/wsdm/
180. Do Good Marketing
181. tl;dr SEO is changing it’s not about optimising your website for search engines, it’s about optimising your business for search engines
182. Kelvin Newman kelvin@brightonseo.com @kelvinnewman