Presentation is loading. Please wait.

Presentation is loading. Please wait.

 Understanding Network Structure Through User Attributes and Behavior.

Similar presentations


Presentation on theme: " Understanding Network Structure Through User Attributes and Behavior."— Presentation transcript:

1  Understanding Network Structure Through User Attributes and Behavior

2 Describe this network

3 What does it mean?  Hints  Built from the photo sharing website Flickr.  On Flickr, photos are labeled with descriptive keywords called tags.  Nodes represent tags and an edge between tags indicate that they were used to describe the same image.  E.g. if an image is tagged with the word “desk” and “keyboard” the network would show a line connecting those two words.  Network is a 1.5 egocentric network of a single tag  What can we say now?

4 Now with content...

5 Connecting Content and Structure  Structural attributes only tell us a little  Must look at data about nodes and edges to really understand what is happening in a network  Node X has high betweenness is only a description of a statistic  Node X has high betweenness, and the data shows he connects a group of people from the US with a group of people from Spain tells what his role is and why it is important.

6 Example Analysis

7  Network is 1.5 egocentric network of a search term on YouTube  Nodes represent videos that match the search term  Links indicate videos share at least one other keyword in common

8 More Data  Search term is “cubs”  Initial thoughts about what you see in the network?

9 Getting into content: Graph Level  Choose a few videos from each cluster and watch them  See what they are about  Look at their keywords

10 Selected nodes in white and black

11 White Nodes’ Keywords  Cubs, CubFans, baseball, Chicago, Please, Stop, Believing  mlb, 2k12, baseball, major, legaue, ronnie, woo, wilckers, wrigley, cubbies, north, side, billy, goat, curse, illinois, ps3, playstiation, cubs  MLB, 12, The Show, MLB 2k12, Diamond Dynasty, Baseball, triple play, world series, home run derby, PS MOVE, Jose Bautista Chicago, Cubs, win, sports, playstation, ps3, ps vita, video game, so real it's it’s unreal  Chicago Cubs, Chicago, Cubs, Wrigley Field, Opening Day, 2011, number one fan, sports fans, baseball, major leagues  Chicago, Cubs, Spring, Training, Baseball, Tony, Campana, Brett, Jackson, Sports, Hohokam, Park, Cactus, League

12 Black Nodes’ Keywords  dog, dogs, puppies, pup, cute, adorable, snuggle, bear cub, Medvjedić, Bär, orsacchiotto, brown bear cub, bears, teddy, medo srečko, cubs, medvedji mladič, slovenia, slovenija  National Geographic, polar, bear, cubs, mother, mom, parent, learn, teach, cute, fluffy, sweet, predator, arctic, predation, hunt  Tiger, Rescue, Lions, Leopards, Cubs, Kittens, Tiger cubs, Wild animal orphanage, Big Cat Rescue, Texas, Tigers, Rescued, Scary, Roar, Rawr, Attack, Aggressive, Sanctuary, Global  tiger, tigress, cubs, machli, fight, nick, ranthamore, croc, crocodile, mugger, india, rajastan, valmik, thapar, bbc, wildlife  cheetah, cheetahs, african, wild, cute, animals, baby, BBC, cubs

13 Conclusions  The cluster of nodes with the white samples represent videos about the baseball team the Chicago Cubs  The cluster of nodes with the black samples represent videos about baby animals (bear cubs, tiger cubs, etc.)

14 Getting into Content: Node Level  Individual nodes may represent different types in a network  This requires understanding node attributes and linking it to the role in the network.

15 Example: Nodes colored by department

16 Example: Detecting User Roles  Study by Welser, Gleave, and Smith, 2007.  Examined the roles users play in discussion groups

17 Example Network

18 Breaking Down Into Egocentric Nets

19 Observations  36 nodes have only one neighbor, and in almost all cases that neighbor has a high degree and had replied to the central node.  Another 17 nodes have two neighbors with this same pattern.  This accounts for nearly 60% of the nodes in the network.  Do these nodes have something in common?

20 Diving Into Content  Group nodes by attributes of their egocentric networks  Look at the behavior of those nodes in discussion groups to see if there are patterns  This involves actually reading their posts and understanding the communication on a content level, not just a network structure level.

21 Findings  Nodes with high out degree and low network density tend to answer a lot of questions, but not engage in a lot of discussion  Nodes with low degrees are generally asking questions. They get a reply and then stop participating.  Many other patterns found by researchers  These results rely on connecting structure to content

22 Conclusions  To understand a network, we need more than structural attributes  Connecting structure with analysis of content can lead to much deeper insights about what is happening in a network.  This is a connection is critical for full, deep, and insightful network analysis.


Download ppt " Understanding Network Structure Through User Attributes and Behavior."

Similar presentations


Ads by Google