unfollowing on twitter<br />FundaKivran-Swaine, PriyaGovindan, Mor NaamanRutgers University | School Media Information Lab...
blue = unfollows<br />
big story<br />online social networks – large proportion of activity on the Web.<br />generate a better understanding of t...
what structural social network properties of nodes and dyads predict the breaking of ties?<br />
theory background<br />tie strength<br />embeddedness within networks<br />power & status<br />
data<br />content data set (911 users) from Naaman , Boase, Lai (2010)social network data from Kwak et al. (2010)<br />715...
analysis<br />* independent variables (computed)<br />	seed properties      	follower-count, follower-to-followee ratio, n...
effect of number of followers (none):<br />how many followers a user has, versus the tendency of people to stop following ...
effect of reciprocity (large):<br />how many “follow” relationships were terminated, when connection was reciprocated, and...
effect of reciprocity (another view):<br />The user’s tendency to reciprocate, versus the tendency of users to stop follow...
effect of follow-back rate:<br />what percentage of people the user follow that follow them back, versus the tendency of p...
effect of common neighbors:<br />the proportion of unfollows amongst pairs with 0,1,2,…,15 common neighbors<br />note: aga...
in-depth analysis<br />*  the details you did not want to know…  <br />*  multi-level logistic regression (dyads/edges nes...
some results<br />… explaining tie-breaks<br />effect of tie strength on breaking of ties.<br />*** dyadic reciprocity 	<b...
some results<br />… explaining tie-breaks<br />effect of power & status on breaking of ties.<br />*** prestige ratio		<br ...
some results<br />… explaining tie-breaks<br />effect of embeddedness on breaking of ties.<br />*** common neighbors		<br />
limitations & future work<br />only two snapshots: add more<br />additional (non-structural) variables (frequency of posti...
for more details<br />http://www.ayman-naaman.net/?p=667<br />FundaKivran-Swaine, PriyaGovindan and Mor Naaman (2011). The...
thank you<br />mornaaman.com<br />@informor<br />
Upcoming SlideShare
Loading in...5
×

Unfollowing on twitter

1,003

Published on

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

  • Be the first to like this

No Downloads
Views
Total Views
1,003
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
4
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Unfollowing on twitter

  1. 1. unfollowing on twitter<br />FundaKivran-Swaine, PriyaGovindan, Mor NaamanRutgers University | School Media Information Lab<br />Article: The Impact of Network Structure on Breaking Ties in Online Social Networks: Unfollowing on Twitter. To Appear, CHI 2011. <br />
  2. 2. blue = unfollows<br />
  3. 3. big story<br />online social networks – large proportion of activity on the Web.<br />generate a better understanding of the dynamics<br />validate theories from social sciences in these new and important settings<br />
  4. 4. what structural social network properties of nodes and dyads predict the breaking of ties?<br />
  5. 5. theory background<br />tie strength<br />embeddedness within networks<br />power & status<br />
  6. 6. data<br />content data set (911 users) from Naaman , Boase, Lai (2010)social network data from Kwak et al. (2010)<br />715 seed nodes<br /> 245,586 “following” connections to seed nodes <br /> 30.6 % dropped between 07/2009 & 04/2010<br />
  7. 7. analysis<br />* independent variables (computed)<br /> seed properties follower-count, follower-to-followee ratio, network density, reciprocity rate, follow-back rate<br />follower propertiesfollower-count, follower-to-followee ratio<br /> dyad propertiesreciprocity, common neighbors, common followers, common friends, right transitivity, left transitivity, mutual transitivity, prestige ratio<br />
  8. 8. effect of number of followers (none):<br />how many followers a user has, versus the tendency of people to stop following that user<br />
  9. 9. effect of reciprocity (large):<br />how many “follow” relationships were terminated, when connection was reciprocated, and when it wasn’t.<br />note: this analysis is not robust due to between-node effects (because we looked at followers of 715 nodes). See paper for a more robust analysis showing the (large) effect of reciprocity.<br />
  10. 10. effect of reciprocity (another view):<br />The user’s tendency to reciprocate, versus the tendency of users to stop following them<br />note: this effect is mostly explained by the individual relationships (previous slide), but included here for dramatic purposes (steep curve!) .<br />
  11. 11. effect of follow-back rate:<br />what percentage of people the user follow that follow them back, versus the tendency of people to stop following the user<br />note: this strong effectsuggests that “follow-back ratio” is a indeed a good measure of status.<br />
  12. 12. effect of common neighbors:<br />the proportion of unfollows amongst pairs with 0,1,2,…,15 common neighbors<br />note: again, this analysis is not robust due to between-node effects (because we looked at followers of 715 nodes). See paper for a more robust analysis showing the effect of common neighbors.<br />
  13. 13. in-depth analysis<br />* the details you did not want to know… <br />* multi-level logistic regression (dyads/edges nested within seed nodes)<br />* three models; full one includes seed, follower, and dyadic/edge variables<br />* full details in the paper<br />
  14. 14. some results<br />… explaining tie-breaks<br />effect of tie strength on breaking of ties.<br />*** dyadic reciprocity <br />*** network density <br />*** highly statistically significant<br />
  15. 15. some results<br />… explaining tie-breaks<br />effect of power & status on breaking of ties.<br />*** prestige ratio <br />*** follow-back rate <br />*** f’s follower-to followee ratio <br />*** dyadic reciprocity <br />*** highly statistically significant<br />
  16. 16. some results<br />… explaining tie-breaks<br />effect of embeddedness on breaking of ties.<br />*** common neighbors <br />
  17. 17. limitations & future work<br />only two snapshots: add more<br />additional (non-structural) variables (frequency of posting?)<br />
  18. 18. for more details<br />http://www.ayman-naaman.net/?p=667<br />FundaKivran-Swaine, PriyaGovindan and Mor Naaman (2011). The Impact of Network Structure on Breaking Ties in Online Social Networks: Unfollowing on Twitter. In Proceedings, CHI 2011.<br />
  19. 19. thank you<br />mornaaman.com<br />@informor<br />
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.

×