Motif‐based embedding label propagation algorithm for community detection

Abstract

Community detection can exhibit the aggregation behavior of complex networks. Network motifs are the fundamental building blocks which can reveal the higher‐order structure of complex networks. Label propagation algorithm has the advantage of approximately linear time complexity, unfortunately, the randomness of label update is a major but unsolved issue. For these reasons, this paper proposes a novel community detection method, named motif‐based embedding label propagation algorithm (MELPA). First, complex network topology is reconstructed by merging higher‐order topology with lower‐order connectivity features, where higher‐order topology is captured by mining network motifs. Second, We design a label propagation characteristic model according to nodes influence, then a new label update rule is formulated based on reconstructed weighted network, the rule integrates frequency among neighbor labels, influence of nodes, propagation characteristics and closeness of nodes to update the node label, the purpose is to overcome the randomness of label selection and identify a better and more stable community structure. Finally, extensive experiments on synthetic networks and real‐world complex networks are conducted to verify the effectiveness of MELPA, especially for the complex networks with unobvious community structure, MELPA will get unexpected results.

GHOC: A generative model for hybrid‐order community detection

“…As a counterpart of lower-order community detection, higher-order community detection has attracted an increasing amount of attention in the past few years. [1][2][3][4][5][26][27][28] One of the earliest attempts of higher-order community detection is motif modularity, 27 which is extended from the Newman-Girvan modularity model by considering the fraction of motif instances within the communities minus the fraction in a random network preserving the same degree of nodes. In Huang et al, 5 a harmonic motif modularity method is designed, which extends motifmodularity into the case of multilayer network consisting of multiple layers.…”

“…In the past few years, motif‐based higher‐order community detection has attracted a large amount of attention in the field of network analysis 1–5 . Compared with the conventional lower‐order community detection only utilizing the lower‐order connectivity pattern which can be captured at the level of individual nodes and edges, 6–10 higher‐order community detection mainly relies on the higher‐order connectivity pattern at the level of small subnetworks, namely motif 1 .…”

VISEL: A visual and magnetic fusion‐based large‐scale indoor localization system with improved high‐precision semantic maps

“…How to enhance the efficiency of fusion strategy? Recent researches 2,13,31 of indoor fusion positioning are loosely coupled and ignore the influence of the environment, locations are directly generated by individual localization modules and then integrated to obtain a fused result. But they rely on an assumption that each individual localization module must be accurate, which is hard to tenable in practical scenes of frequent occlusion and signal mutation.…”

“…With the popularity of wireless networks and smartphones, the positioning methods based on WiFi, 8,9 Bluetooth, 10,11 and Pedestrian Dead Reckoning (PDR) 12 have great potential for mobile computing and universal application, which has been developed well. Single signal‐based localization algorithms, however, suffer from signal fingerprint ambiguities and long‐term maintenance of the database, 13 the localization error in large‐scale indoor scenes is relatively large 14 . Multisource fusion localization has become the mainstream method in large‐scale indoor scenes 15–17 .…”

An improved integral distinguisher scheme based on neural networks

“…Novel machine learning architectures have made a great breakthrough in various fields, including drone navigation, 1 speech emotion recognition, 2 and recommendation systems 3 . Other applications include but are not limited to estimating the price of crude oil, 4 detection of solder paste, 5,6 modeling rumor spreading, 7 community detection, 8 and so forth 9–12 . They can be trained using different methods, such as federated learning 13,14 and collaborative learning 15 .…”