Home
This Title All WIREs
WIREs RSS Feed
How to cite this WIREs title:
WIREs Data Mining Knowl Discov
Impact Factor: 2.541

Predicting the dynamics of social circles in ego networks using pattern analysis and GA K‐means clustering

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

The tremendous amount of content generated on online social networks has led to a radical paradigm shift in the interest of people to group friends dynamically and share content selectively. At large, social networking sites (e.g. Google+, Facebook, Twitter, etc.) offer users with various controls over categorizing their family members, friends, colleagues, etc. into one or more ‘circles’ that they want to share content with. However, it is typically impossible to design social circles in large networks and update their number and size, whenever networks grow. Aiming at predicting the dynamics of formation and evolution of social circles, we performed several experiments on ground‐truth data, and found that studying patterns of network and profile features at individual level rather than studying circle as a whole can greatly enhance the understanding of social circles development in online social networks. In this review, we first present a comprehensive study of the structural behavior of circles, and then build an observation that within every circle there exist some key elements, termed as ‘Node of Creations (NoCs)’, which play an important role in the development, survival, and evolvability of circle structures. We, therefore, propose a Genetic Algorithm–based framework to determine these key elements (NoCs) and differentiate Ego networks into non‐overlapping, hierarchically nested as well as overlapping circles by leveraging knowledge from the identified patterns in order to assist K‐means clustering. We have performed our experiments using Facebook and Twitter datasets and the experimental results clearly demonstrate the effectiveness of our scheme. WIREs Data Mining Knowl Discov 2015, 5:113–141. doi: 10.1002/widm.1150 This article is categorized under: Technologies > Machine Learning Technologies > Prediction Technologies > Structure Discovery and Clustering
Computation of parameters for user 584 in circle 1. (a) Strength of ties between user 584 and core group members. (b) Profile similarity between user 584 and core group members. (c) Strength of ties between user 584 and residual group members. (d) Profile similarity between user 584 and residual group members.
[ Normal View | Magnified View ]
Representation of core (C) and residual area (R) for circle 1.
[ Normal View | Magnified View ]
An example ESN.
[ Normal View | Magnified View ]
Model structure of a circle in an ESN.
[ Normal View | Magnified View ]
Comparisons of the proposed CDCRes components schemes (i.e. Case I, Case II, and Case III) with the proposed hybrid scheme on 15 ego networks.
[ Normal View | Magnified View ]
Performance of detecting ground‐truth social circles on 15 ego networks from Twitter using different measures: Precision, Recall, F1‐score, and F2‐score.
[ Normal View | Magnified View ]
Performance of detecting ground‐truth social circles on 10 ego networks from Facebook using different measures: Precision, Recall, F1‐score, and F2‐score.
[ Normal View | Magnified View ]
BER of detecting ground‐truth social circles on 50 ego networks from Twitter.
[ Normal View | Magnified View ]
BER of detecting ground‐truth social circles on 10 ego networks from Facebook.
[ Normal View | Magnified View ]
Prediction accuracy of detecting ground‐truth social circles on 50 ego networks from Twitter.
[ Normal View | Magnified View ]
Prediction accuracy of detecting ground‐truth social circles on 10 ego networks from Facebook.
[ Normal View | Magnified View ]
Fitness graphs showing error bars across 100 generations for 10 runs. (a) Best fitness scores and (b) average fitness scores.
[ Normal View | Magnified View ]
Representation of flexible modeling of social circles in an ESN using CDCRes: (a) non‐overlapping, (b) overlapping, and (c) hierarchically nested.
[ Normal View | Magnified View ]
Comparison between the predicted and ground truth circles in example ESN.
[ Normal View | Magnified View ]
Representation of individuals' behaviors using four different circles: for example Figure (a), (c), and (e) illustrate variations in deg_cen, str and prof_sim of every individual belonging to circle 1 with the remaining members of the same circle and Figure (b), (d), and (f) also provide values computed by the same metrics but with the members existing outside circle 1. Similar representation is given in Figures for circle 2, circle 3, and circle 4, respectively.
[ Normal View | Magnified View ]
Distribution of density (a,b) and profile (c,d) similarity of circles on Facebook and Twitter
[ Normal View | Magnified View ]
Structure of circles in an ego network from Facebook. Within a circle, each alter is connected with the other alters of the same circle, and can as well have social connections outside the circle. Within a circle, connectivity differs from alter to alter. Encircled alters are representing the nodes with highest connectivity and act as node of creations (NoCs) in their respective circles.
[ Normal View | Magnified View ]

Browse by Topic

Technologies > Machine Learning
Technologies > Prediction
Technologies > Structure Discovery and Clustering

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts