null-models

A python3 module for maximum entropy models on networks.

Randomization of presence/absence species distribution raster data for calculating standardized effect sizes and testing null hypothesis.

Introduction to basic network measurements and working with null models in Networkx.

Null Models for Labeled Multi-graphs

Reference implementation for separating graph dependence from graph advantage in graph-ablation studies.

Calibration and falsification toolkit for AI claims: null models, evidence grades, claim boundaries and observer-aware workflow examples.

Reproducible structural benchmark of Voynich Manuscript regularities; no decipherment claim

MCMC algorithms to sample random bipartite graphs with given left and right degree sequences and BJDM.

Functions to perform the restricted null model described in Felix et al 2017 (DOI: 10.1101/236687).

Code for trait based simulations of network structure and case study

The methodology arm of the Arithmon program: how surprising is a claimed exact relation between mathematical invariants and measured physical constants? Frozen inputs deposited with a dated DOI, three expression grammars, four null models, calibration on historical verdicts, and a scorecard reusable on any framework.

Commented codes for the IHS model.

Ten interpretable structural features per node, from the graph alone, with a measured applicability domain

Maximum entropy null models for bipartite networks

How to create randomized matrices in R for analyzing in Pajek.

Null-model test for modern mammal bioregionalization: β-sim dissimilarity, UPGMA clustering, and a curveball randomization null, validated end-to-end on synthetic data.

Analysis of temporal fluctuations in Stack Overflow user activity to characterize interaction-network structure, test fluctuation-based null models, and infer candidate connections through lagged covariance and inverse-network methods, developed for HUJI’s Dynamics, Networks and Computation course (67562).

A preregistered, algorithm-agnostic matched-null test for cluster-count claims ("N types"), tried out on personality.

Functional diversity of an urban campus under construction (Las Peñas, ESPOL, Guayaquil). R pipeline computing FRic, FEve, FDiv, FDis, Rao's Q and functional redundancy on Gower distances with Cailliez-corrected PCoA, plus null models, taxonomic contrast and abundance-weighting sensitivity. Flora and fauna analysed separately by design.