Mixture models with entropy regularization for community detection in networks

Chang, Zhenhai; Yin, Xuefeng; Jia, Caiyan; Wang, Xiaoyang

doi:10.1016/j.physa.2018.01.002

Cited by 8 publications

(5 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In external evaluation, the comparison algorithm is AGDL 29 , Fluid 30 , Belief 31 , CPM 32 , Chinese Whispers 33 , DER 34 , Eigenvector 35 , EM 36 , Genetics Algorithm 14 , Girvan Newman 12 , Greedy Modularity 13 , Kcut 37 , Label Propagation 38 , Leiden 11 , Louvain 10 , Markov Clustering 39 , RBER Pots, RB Pots 40 [33], Significance 41 , Spinglass 40 , Surprise 42 ,Walktrap 43 , Head tail 44 , LSWL+ 45 , Paris , dan Regularized spectral 46 using the library CDLIB 47,48 dan Networkx 49 .…”

Section: Discussionmentioning

confidence: 99%

Community detection with greedy modularity disassembly strategy

Rustamaji¹,

Kusuma²,

Nurdiati³

et al. 2023

Preprint

View full text Add to dashboard Cite

One of the problems in networking is community detection, the process of identifying groups of nodes in a network that are more densely connected than the rest of the network. To measure the quality of community detection in the network using modularity. Greedy modularity is one of the community detection algorithms. It optimizes the modularity score locally at each step rather than globally optimizing over all possible network partitions. However, because the algorithm makes local improvements, it can be trapped in a suboptimal network partition. In this research, we developed the Greedy Modularity algorithm by adding an exploration method: a disassembly node and disassembly community formed to increase modularity. There are four strategies for disassembly nodes: disassembly random nodes, weak nodes, nodes with low embeddedness values, and releasing nodes that do not form triads. The disassembly community strategy includes random communities, weak communities, communities with low internal edge density, communities with low triad participation ratio, and communities with low conductance. The results show increased modularity in various real-world and synthetic network datasets. We also compared the results with various community detection algorithms, and the modularity ranking of these algorithms were in positions 1-8 across various datasets.

show abstract

Section: Discussionmentioning

confidence: 99%

Community detection with greedy modularity disassembly strategy

Rustamaji¹,

Kusuma²,

Nurdiati³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Values close to 1 indicate a strong correlation between two variables, whereas values close to 0 indicate a weak correlation. For the external evaluation, the comparison algorithms were AGDL 31 , Fluid 32 , Belief 33 , CPM 34 , Chinese Whispers 35 , DER 36 , Eigenvector 37 , EM 38 , Genetics Algorithm 16 , Girvan Newman 14 , Greedy Modularity 15 , Kcut 39 , Label Propagation 40 , Leiden 13 , Louvain 12 , Markov Clustering 41 , RBER Pots, RB Pots 42 [33], Significance 43 , Spinglass 42 , Surprise 44 , Walktrap 45 , Head tail 46 , LSWL+ 47 , Paris, dan Regularized spectral 48 using the library CDLIB 49 , 50 dan Networkx 51 .…”

Section: Methodsmentioning

confidence: 99%

Community detection with Greedy Modularity disassembly strategy

Rustamaji,

Kusuma,

Nurdiati

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Community detection recognizes groups of densely connected nodes across networks, one of the fundamental procedures in network analysis. This research boosts the standard but locally optimized Greedy Modularity algorithm for community detection. We introduce innovative exploration techniques that include a variety of node and community disassembly strategies. These strategies include methods like non-triad creating, feeble, random as well as inadequate embeddedness for nodes, as well as low internal edge density, low triad participation ratio, weak, low conductance as well as random tactics for communities. We present a methodology that showcases the improvement in modularity across the wide variety of real-world and synthetic networks over the standard approaches. A detailed comparison against other well-known community detection algorithms further illustrates the better performance of our improved method. This study not only optimizes the process of community detection but also broadens the scope for a more nuanced and effective network analysis that may pave the way for more insights as to the dynamism and structures of its functioning by effectively addressing and overcoming the limitations that are inherently attached with the existing community detection algorithms.

show abstract

“…It also reflects the complexity of the sequence, and its value is proportional to the complexity. Compared with approximate entropy, sample entropy has the following advantages: (1) it does not depend on the length of the data; (2) it has better consistency [ 27 ]. The sample entropy

can be represented as:

…”

Section: Methodsmentioning

confidence: 99%

Cost-Sensitive KNN Algorithm for Cancer Prediction Based on Entropy Analysis

Song

2022

Entropy

View full text Add to dashboard Cite

Early diagnosis of cancer is beneficial in the formulation of the best treatment plan; it can improve the survival rate and the quality of patient life. However, imaging detection and needle biopsy usually used not only find it difficult to effectively diagnose tumors at early stage, but also do great harm to the human body. Since the changes in a patient’s health status will cause changes in blood protein indexes, if cancer can be diagnosed by the changes in blood indexes in the early stage of cancer, it can not only conveniently track and detect the treatment process of cancer, but can also reduce the pain of patients and reduce the costs. In this paper, 39 serum protein markers were taken as research objects. The difference of the entropies of serum protein marker sequences in different types of patients was analyzed, and based on this, a cost-sensitive analysis model was established for the purpose of improving the accuracy of cancer recognition. The results showed that there were significant differences in entropy of different cancer patients, and the complexity of serum protein markers in normal people was higher than that in cancer patients. Although the dataset was rather imbalanced, containing 897 instances, including 799 normal instances, 44 liver cancer instances, and 54 ovarian cancer instances, the accuracy of our model still reached 95.21%. Other evaluation indicators were also stable and satisfactory; precision, recall, F1 and AUC reach 0.807, 0.833, 0.819 and 0.92, respectively. This study has certain theoretical and practical significance for cancer prediction and clinical application and can also provide a research basis for the intelligent medical treatment.

show abstract

Mixture models with entropy regularization for community detection in networks

Cited by 8 publications

References 36 publications

Community detection with greedy modularity disassembly strategy

Community detection with greedy modularity disassembly strategy

Community detection with Greedy Modularity disassembly strategy

Cost-Sensitive KNN Algorithm for Cancer Prediction Based on Entropy Analysis

Contact Info

Product

Resources

About