Skip to main content

Frontiers of Engineering Reports on Leading-Edge Engineering from the 2013 Symposium (2014) / Chapter Skim
Currently Skimming:

Modeling Large-Scale Networks--Tony Jebara
Pages 5-10

The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.


From page 5...
... For example, in a class of ­ incoming freshmen students with no known friendship connections, is it possible to predict which pairs will become friends at the end of the year using only their demographic profile information? Similarly, can ­ olocation be used to predict c communication -- that is, using only the location history of a population of mobile phone users, can an observer predict which pairs of users are likely to communicate with each other?
Read the entire page →
From page 6...
... More specifically, A and X permit the estimation of an appropriate distance metric that can then be used to compute the distance between user i's demographic vector xi and user j's demographic vector xj. This distance metric helps quantify how various attributes compete in the formation of friendship links, for example, how much does an age difference matter relative to a height difference when computing the affinity or distance between a pair of users?
Read the entire page →
From page 7...
... For instance, through Facebook social networks it appears that when students form friendships, those at Harvard are relatively more attentive to differences in relationship status, those at Stanford and Columbia are relatively more sensitive to differences in graduation year, and for those at MIT differences in dorm assignments are most important. Thus, social network structure helps tease apart demographic attributes and determine which are more or less relevant.
Read the entire page →
From page 8...
... Note that the relative importance of the demographic attri­ utes varies across the four different universities in the study. b Jebara Figure 2_R02544.eps likely form between new users and elucidate which specific aspects of user profiles and demographics are predictive of communication and social interaction (Lazer et al.
Read the entire page →
From page 9...
... Proceedings of the International Conference on Machine Learning (ICML) , Montreal, June.
Read the entire page →

Key Terms

  • demographic attributes
  • mobile phone
  • connectivity information
  • social networks
  • social network

  • facebook data
  • mahalanobis distance
  • distance metrics
  • neural information
  • profile information

  • call detail
  • location history
  • information processing
  • training data
  • relationship status

  • detail records
  • estimate distance
  • euclidean distance
  • appropriate distance
  • malahanobis distance

  • mobile data
  • correct connectivity
  • static demographic
  • reconstructing facebook
  • network helps

  • multivariate demographic
  • metric helps
  • apart demographic
  • demographic vector
  • demographic profile


    • This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
    More information on Chapter Skim is available.