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) .

Spectral Community Detection for Sparse Networks

Implements spectral clustering algorithms for community detection in sparse networks under the stochastic block model ('SBM') and degree-corrected stochastic block model ('DCSBM'), following the methods of Lei and Rinaldo (2015) . Provides a regularized normalized Laplacian embedding, spherical k-median clustering for 'DCSBM', standard k-means for 'SBM', simulation utilities for both models, and a misclustering rate evaluation metric. Also includes the 'NCAA' college football network of Girvan and Newman (2002) as a benchmark dataset, and the Bethe-Hessian community number estimator of Hwang (2023) .

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 Detection by Scaled Null-Adjusted Persistence

Finds the vertex partition of an undirected graph that maximises a persistence-based objective, using the Milano local-search algorithm. Three related quality measures are supported: the persistence probability of a community, its null-adjusted persistence (NAP), and the scaled null-adjusted persistence (Scaled-NAP), which interpolates between NAP and modularity. The methods are described in Avellone et al. (2023) and Avellone et al. (2025) .

Inductive Node-Splitting Cross-Validation for Community Detection

Implements Inductive Node-Splitting Cross-Validation (INCV) for selecting the number of communities in stochastic block models. Provides f-fold and random-split node-level cross-validation, along with competing methods including CROISSANT, Edge Cross-Validation (ECV), and Node Cross-Validation (NCV). Supports both SBM and Degree-Corrected Block Models (DCBM), with multiple loss functions (L2, binomial deviance, AUC). Includes network simulation utilities for SBM, RDPG, and latent space models.

Graph Community Detection Methods into Systematic Conservation Planning

An innovative tool-set that incorporates graph community detection methods into systematic conservation planning. It is designed to enhance spatial prioritization by focusing on the protection of areas with high ecological connectivity. Unlike traditional approaches that prioritize individual planning units, 'priorCON' focuses on clusters of features that exhibit strong ecological linkages. The 'priorCON' package is built upon the 'prioritizr' package , using commercial and open-source exact algorithm solvers that ensure optimal solutions to prioritization problems.

"Node+Edge Betweenness Community Detection Algorithm for 'igraph' Objects"

Implements the "Node + Edge Betweenness" community detection algorithm for network analysis using 'igraph' objects. This algorithm combines node degree and betweenness centrality measures to identify communities within networks, with a gradient evident in social partitioning. The package provides functions for implementation of the algorithm, visualization, and analysis of the resulting community structure along with the original dataset used in the publication. Methods are based on results from Smith, Pittman and Xu (2026) .