Network Analysis and Community Detection

Features tools for the network data analysis and community detection. Provides multiple methods for fitting, model selection and goodness-of-fit testing in degree-corrected stochastic blocks models. Most of the computations are fast and scalable for sparse networks, esp. for Poisson versions of the models. Implements the following: Amini, Chen, Bickel and Levina (2013) Bickel and Sarkar (2015) Lei (2016) Wang and Bickel (2017) Zhang and Amini (2020) Le and Levina (2022) .

Network-Adjusted Covariates for Community Detection

Incorporating node-level covariates for community detection has gained increasing attention these years. This package provides the function for implementing the novel community detection algorithm known as Network-Adjusted Covariates for Community Detection (NAC), which is designed to detect latent community structure in graphs with node-level information, i.e., covariates. This algorithm can handle models such as the degree-corrected stochastic block model (DCSBM) with covariates. NAC specifically addresses the discrepancy between the community structure inferred from the adjacency information and the community structure inferred from the covariates information. For more detailed information, please refer to the reference paper: Yaofang Hu and Wanjie Wang (2023) . In addition to NAC, this package includes several other existing community detection algorithms that are compared to NAC in the reference paper. These algorithms are Spectral Clustering On Ratios-of Eigenvectors (SCORE), network-based regularized spectral clustering (Net-based), covariate-based spectral clustering (Cov-based), covariate-assisted spectral clustering (CAclustering) and semidefinite programming (SDP).

Hierarchical Community Detection by Recursive Partitioning

Hierarchical community detection on networks by a recursive spectral partitioning strategy, which is shown to be effective and efficient in Li, Lei, Bhattacharyya, Sarkar, Bickel, and Levina (2018) . The package also includes a data generating function for a binary tree stochastic block model, a special case of stochastic block model that admits hierarchy between communities.

Dynamic Network Communities Detection and Generation

Used for evolving network analysis regarding community detection. Implements several algorithms that calculate communities for graphs whose nodes and edges change over time. Edges, which can have new nodes, can be added or deleted. Changes in the communities are calculated without recalculating communities for the entire graph. REFERENCE: M. Cordeiro et al. (2016) G. Rossetti et al. (2017) G. Rossetti (2017) R. Sarmento (2019) .

Differential Community Detection in Paired Biological Networks

A differential community detection (DCD) based approach to effectively locate differential sub-networks in paired scale-free biological networks, e.g. case vs control - Raghvendra Mall et al (2017) .

Community Structure Detection via Modularity Maximization

The algorithms implemented here are used to detect the community structure of a network. These algorithms follow different approaches, but are all based on the concept of modularity maximization.

Weighted Fast Greedy Algorithm

Implementation of Weighted Fast Greedy algorithm for community detection in networks with mixed types of attributes.

ROBustness in Network

Assesses the robustness of the community structure of a network found by one or more community detection algorithm to give indications about their reliability. It detects if the community structure found by a set of algorithms is statistically significant and compares the different selected detection algorithms on the same network. robin helps to choose among different community detection algorithms the one that better fits the network of interest. Reference in Policastro V., Righelli D., Carissimo A., Cutillo L., De Feis I. (2021) < https://journal.r-project.org/archive/2021/RJ-2021-040/index.html>.

Covariate Assisted Spectral Clustering on Ratios of Eigenvectors

Functions for implementing the novel algorithm CASCORE, which is designed to detect latent community structure in graphs with node covariates. This algorithm can handle models such as the covariate-assisted degree corrected stochastic block model (CADCSBM). CASCORE specifically addresses the disagreement between the community structure inferred from the adjacency information and the community structure inferred from the covariate information. For more detailed information, please refer to the reference paper: Yaofang Hu and Wanjie Wang (2022) . In addition to CASCORE, this package includes several classical community detection algorithms that are compared to CASCORE in our paper. These algorithms are: Spectral Clustering On Ratios-of Eigenvectors (SCORE), normalized PCA, ordinary PCA, network-based clustering, covariates-based clustering and covariate-assisted spectral clustering (CASC). By providing these additional algorithms, the package enables users to compare their performance with CASCORE in community detection tasks.

Clique Percolation for Networks

Clique percolation community detection for weighted and unweighted networks as well as threshold and plotting functions. For more information see Farkas et al. (2007) and Palla et al. (2005) .