Learning to Discover Social Circles in Ego Networks

By Julian McAuley, Jure Leskovec. 2012

2,296 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Representation Learning on Graphs with Jumping Knowledge Networks (2018, methodology): “As an example, consider the social network in Figure 1 from GooglePlus (Leskovec & Mcauley, 2012).”
struc2vec: Learning Node Representations from Structural Identity (2017, methodology): “We apply the edge sampling model to an egonet extracted from Facebook (224 nodes, 3192 edges, max degree 99, min degree 1) [8] to generate G1 and G2 with dierent values for s .”
Decentralized Federated Learning: Fundamentals, State of the Art, Frameworks, Trends, and Challenges (2022, methodology): “[73] 2022 Decentralized recommender system based on the principles of gossip learning Cross-device Decentralized Simulation GMF PRME-G FedAvg FedFast Reptile Synchronous Scheduling Foursqueare-NYC [178] Gowalla-NYC [179] MovieLens [180] [74] 2022 Gossip-based system to maximize social awareness amon”
Evolutionary dynamics on any population structure (2016, methodology): “peration. e, Table showing data from panels a{d as well as a social network of family, self-reported friends, and coworkers as of 1971 from the Framingham Heart Study [43, 44], a Facebook ego-network [45], and the co-authorship network for the General Relativity and Quantum Cosmology category of the”
Overlapping community detection at scale: a nonnegative matrix factorization approach (2013, background): “In terms of scalability most overlapping community detection methods scale to networks with at most thousands of nodes [2, 22, 25].”
SNAP: A General Purpose Network Analysis and Graph Mining Library (2016, methodology): “ [Yang et al. 2014]. Community-Afﬁliation Graph Model identiﬁes communities in the entire network. SNAP also provides a complementary approach to network wide community detection. The Circles method [McAuley and Leskovec 2012] uses the friendship network connections as well as user proﬁle informatio”
Predictive Network Representation Learning for Link Prediction (2017, background): “Facebook [4] is a sample of online social network, where nodes represent users and edges represent friendships.”
Detecting cohesive and 2-mode communities indirected and undirected networks (2014, methodology): “, social circle) formation: Facebook is an undirected network and in Facebook circles are driven by dense mutual friendships among users with homogeneous backgrounds [31]; therefore, we would expect cohesive communities in Facebook.”
Community Detection in Networks with Node Attributes (2013, methodology): “Finally, we consider theCircles [28] method, which represents overlapping hard-membership approaches.”
User Profile Preserving Social Network Embedding (2017, methodology): “Each node is described by a 940-dimensional bag-of-words vector constructed from treestructured user profiles [Leskovec and Mcauley, 2012].”
Data Poisoning Attack against Unsupervised Node Embedding Methods (2018, background): “Facebook (Leskovec & Mcauley, 2012): a social networks with 4039 nodes and 88234 edges.”
Querying k-truss community in large and dynamic graphs (2014, methodology): “Furthermore, as the communities a user participates in represent the social contexts of the user, online community search provides a useful tool for other analytical tasks, such as social circle discovery [14] and social contagion modeling [20].”
Spanning Tree Based Community Detection Using Min-Max Modularity (2016, background): “Tneigh(u, v) = |neighbour(u) ∪ neighbour(v)| (13) Step 3: After calculating weight of each edge in step 2, all the edges may be arranged in nondecreasing order.”
A new space for comparing graphs (2014, methodology): “tures from random Erdos-Reyni graphs. We expect methodologies which capture properties like triadic closure and the degree distribution to perform well on this task. We used the Twitter6 ego networks [20], which is a large public dataset of social ego networks, which contains around 950 ego networks”