Questions tagged [networks]

Refers to network theory as part of the graph theory. For questions about neural networks, use our [neural-networks] tag.

Filter by
Sorted by
Tagged with
0
votes
0answers
15 views

Appropriate method for dynamic clustering of network data in R?

I've got a large set of Twitter data, and I'm trying to show the merging/formation of user communities over time based on hashtag usage. For instance, two users that use the same hashtag in the same ...
1
vote
0answers
11 views

Symbol interpretation in non-constant power law density

I am reading Barabasi book on network science. I am struggling to interpret the density formula for a power-law degree distribution with low-degree saturation and high-degree cut-off (http://...
0
votes
0answers
9 views

How do I know which version of a factor analysis is the correct one?

I am doing network analysis with the bootnet package in R. I'm running into an issue where the way that I compute my network results in a very different structure, and I'm having trouble figuring out ...
0
votes
0answers
15 views

How do I know if violations of assumption of sparsity is problematic?

I am running a network analysis in R using the estimateNetwork(default = "EBICglasso") function from the bootnet package. I have set ...
3
votes
1answer
336 views

How can this CDF be decreasing? (power-law)

I am reading the first vignette of the powerRlaw R package. It uses the moby dataset, which is an array of word frequency. Each element is a word, and the value if ...
0
votes
0answers
12 views

Standard Error of ERGM Coefficients

I am trying to calculate the standard error of ERGM coefficients, which is estimated by MCMC sample. For an ERGM $P(y;\eta) = \exp[\eta^\top g(y) - \psi(\eta)]$, denote $\eta$ as the true parameter, $\...
0
votes
0answers
8 views

What a log-log plot means in the context of a real world network?

I've collected a directed network from Twitter that consists of about 25,000 nodes and 850,000 edges. After plotting the degree distribution on a log-log plot, I have the following image which I am ...
0
votes
1answer
69 views

Network analysis: density of communities/partitions (and other metrics)

After running a community detection algorhythm (e.g. edge betweenness, or greedy modularity), I like know the density of each seperate community, and potentially some other metrics, too. My desired ...
1
vote
0answers
12 views

what would a network looks like if the modularity is almost -1?

I know that larger positive modularity is good, and often lies between -0.5 and 1. But what would the network looks like if the modularity is appoaching -1? For example, equals to -0.7?
1
vote
1answer
11 views

Constructing a Weighted Random Graph

I want to create a weighted random graph (in contrast to the unweighted Erdős–Rényi model). I have a list of weights (derived from a real-world network, very skewed distribution that most weights are ...
1
vote
1answer
9 views

Model Comparison for Weighted Network

I have a weighted network data and I want to know whether this network possesses small-world features or the features of a scale-free network. I know for unweighted networks, I can create comparable ...
1
vote
0answers
22 views

Is Kendall's tau adequate to test for a non-monotonic relationship?

Goal Networks can have so many nodes that it's hard to study them. That is why in some cases nodes are clustered together. I am looking at whether the betweenness of a node is maintained after the ...
1
vote
0answers
23 views

How can I perform a meta-analysis on Kendall's tau?

Goal Networks can have so many nodes that it's hard to study them. That is why in some cases nodes are clustered together. I am looking at whether the betweenness of a node similar to the one of its ...
1
vote
1answer
23 views

Given a graph, test if some vertices are more connected than the background

I have a neighborhood graph of some 10k vertices (a k-NN graph of single-cell RNA-seq data). I am interested if a given set of vertices is more connected to each other than you would expect by chance. ...
0
votes
0answers
15 views

Fixed number of communities using community detection algorithms in r

I have a graph G with 10000 nodes and 30000 edges. I want to partition the graph into two communities only (i.e., Community A and Community B). Criteria: First, we have to make a subgraph of 1000 ...
2
votes
2answers
35 views

Network clustering stability using bootstrapping techniques

I use the standard modularity-maximisation Louvain clustering method to partition a large undirected network into communities. I fear the result of the partition is quite fragile. Is there a standard ...
0
votes
0answers
37 views

Small-worldness for Weighted Undirected Graph in R

As an input, I have a 28x28 adjacency matrix containing weights (i.e., strengths of associations among the 28 nodes) using which I create a graph in R, packages: ...
0
votes
0answers
26 views

What is the best approach to examine the dimensionality of related behaviours and to develop theory?

I’m preparing a study whose main goal it is to explore whether a set of related behaviours are best conceptualized as one-dimensional or two-dimensional. Traditionally, such questions have been ...
0
votes
0answers
31 views

How to add attributes to nodelist from calculations on an edgelist?

I have a bimodal network of people who write letters soliciting money from various entities made from an edgelist and nodelist (MRE code below). How do I add columns to my nodelist based on ...
1
vote
0answers
23 views

Inference in a system in which (almost) all regressors are endogenous

We conducted a number of tests on individual species over a period of several days. Let $x_{k,s,t}$ denote the result of test $k$ for species $s$ at time $t$. In total there are $S$ species, $K$ tests,...
3
votes
1answer
50 views

Predictions in a system in which (almost) all regressors are endogenous

Consider the following system where all variables are endogenous. \begin{align*} x_{1} & =\beta_{21}x_{2}+\beta_{31}x_{3}+u_{1}\\ x_{2} & =\beta_{12}x_{1}+\beta_{32}x_{3}+u_{2}\\ x_{3} &...
2
votes
1answer
19 views

social network measure of node's crowdedness

The undirected network below shows the interaction of workers (blue nodes) with products worked on during a time period. The thickness of the edges/arcs indicates how much time the worker spends on a ...
0
votes
0answers
14 views

How do I weight many-to-many relationships in a regression model?

In this context, I have 3 types of entities used to build a predictive model: subject - An individual. event - An activity performed by an individual at a particular time. target - An outcome that ...
0
votes
1answer
31 views

Evaluation of gene regulatory network

I am new to this forum and data science, I will try my best to describe my question. Question background: I am molecular biology student and have recently inferred a gene expression network using 105 ...
1
vote
0answers
96 views

How to create a Bayesian network? [closed]

I have the following question and I really need help please: Consider the following hypothetical scenario. A car company would like to use a Bayesian Network model to better predict whether a certain ...
0
votes
0answers
65 views

What is the difference between brokerage and betweenness?

This link describes betweenness: "Betweenness centrality measures the extent to which a vertex lies on paths between other vertices." This link describes brokerage: "Brokerage is a ...
0
votes
0answers
34 views

To estimate a Markov random field, do we have to assume a DAG generated the data?

Even if I believe that I understood the Markov Random Fields and DAGs separately, I encountered a question that I have written in the title and cannot come up with a clear-cut answer. Can you help me ...
0
votes
0answers
8 views

Two different formulations of PageRank

I recently read a paper (requires institutional access) which formulates PageRank as an optimization problem $$ \min_\mathbf{r} d (\mathbf{I}-\mathbf{A})\mathbf{r}+(1-d)\Vert \mathbf{r}-\mathbf{e}\...
0
votes
1answer
22 views

Identify individual networks within dataset containing many (using SAS preferably)

I have a dataset of links/edges that looks something like the following (except significantly larger - approx. 50,000 links): ...
1
vote
0answers
22 views

Estimating a network-like structure with lead/lag relations

This is a very general question, but any ideas or hints towards the right direction would be very helpful. Assume a dataset with a number of countries $i = 1, \ldots, N$ and a set of indicators $X \in ...
0
votes
1answer
237 views

What is the difference between local efficiency and betweenness centrality in network analysis?

There are many network properties that you can extract for a given node. Two I have encountered that sound conceptually very similar to me and that I have difficulty distinguishing are betweenness ...
0
votes
0answers
37 views

Similarity matrix for directed graph vs. undirected graph producing odd/disparate numbers on diagonal (i.e. self-similarity)

In a correlation, distance, or similarity matrix, comparisons to self should be along the diagonal as a constant 1 or 0. However, I have found that while similarity to self is a constant 0 in ...
1
vote
0answers
11 views

Accounting for connectedness during the analysis of nodes in a directed acyclcic graph

I am studying a river network and environmental features measured at different stations along the network. Often, one station is higher up the river than another (or multiple others). I can ...
4
votes
1answer
101 views

What are the conditions for a graph's adjacency matrix to not have a negative eigenvalue with magnitude>=1?

Say I have a (directed) graph $G$ with an adjacency matrix $A$. For the sake of the question, let's assume it's normalized column-wise (edge weights are normalized so the sum of out-edge weights per ...
0
votes
0answers
14 views

implementation of network autocorrelation models

I've googled this but can't find an answer: where can I find a good implementation of a network autocorrelation model. I want to predict a dichotomous response variable from a network independent ...
0
votes
0answers
41 views

Hierarchical graph clustering using a kernel matrix in R

I have a set of 9 directed graphs of differing sizes and I'd like to use graph clustering to create a dendrogram illustrating their structural similarity, similar to what's done in the NetConfer ...
1
vote
0answers
11 views

Methods to align/reduce distance between two or more 3D graph-spaces?

I'm trying to align multiple graphs of spatial data, similar to image registration, but on a 3D graphspace of a select feature. I have a bunch of point data with x,y,z coordinates that together, ...
2
votes
2answers
100 views

How do I evaluate correlation, that appears non-linear

Directed here from StackOverflow Let's say I want to assess if there is a correlation between two fields, one of which I know to have a power distribution. A lot of the information I read assumes ...
2
votes
0answers
27 views

Is a grid-like network more efficient than a random network?

Imagine two different types of street networks with the same number of nodes and where edges are weighted according to their length. Both network types cover the same geographic area (say 1km²) and ...
0
votes
0answers
20 views

Binwidths/no. of bins for 2 histograms for overlaying them

Should the binwidths for two histograms be same or no. of bins in both histograms be same when they are overlayed for comparison? or there is no correlation? I am comparing histogram of centrality for ...
0
votes
1answer
10 views

Network linkage formation prediction

Is there a statistical model that can study determinants of network linkage formation? For a set of companies (where we observe their industry, annual revenue, etc), we see which pairs are connected ...
2
votes
1answer
92 views

Degree and weight preserving randomisation in networks

I have a network in which nodes are highly interconnected (250 nodes where 90% of the nodes have degree = 249). The connections are weighted with a normalised index that goes from 0 to 1, where 1 ...
1
vote
1answer
20 views

How can I encode a network into a machine learning feature?

I am developing a predictive machine learning model. The dataset is social media posts from Twitter relating to a particular topic. One of the features I would like to incorporate into the model is ...
2
votes
0answers
61 views

Intuition behind null model for disparity filter algorithm

I'm trying to better understand the disparity filter algorithm for weighted networks: Original paper: https://www.pnas.org/content/106/16/6483 Wikipedia link: https://en.wikipedia.org/wiki/...
1
vote
1answer
64 views

ERGM with larger log-likelihood but worse fit than reference model?

I'm using the R package statnet to fit some ERGMs to the Faux Dixon High simulated network data provided with the package. The first model I fit is almost identical ...
1
vote
0answers
9 views

What is constraint and effectivesize in structural holes?

Today, I spent the whole day trying to understand what structural holes in network means. More specifically I want to clarify what the two measurements constraint and effective size means? Is it good ...
0
votes
0answers
94 views

Why is my correlation matrix dropping that many NA?

I am trying to build a correlation matrix among documents per topic on a Latent Dirichlet Allocation model by text2vec, getting a doc_topic_distr matrix like below, with only first 5 documents, it's a ...
1
vote
0answers
49 views

Compare networks, including centrality measures, clustering/community detection results

Hi my fellow researchers, I am trying to compare two networks based on exactly same group of individuals (179 unique individuals): One network is based on these individuals' meeting data (two-mode/...
1
vote
0answers
119 views

What is the difference between node clustering and network community detection?

Node clustering algorithms in which we attempt to determine dense regions of the graph based on edge behaviour(either be a distance value or a similarity value). The more similar nodes come together ...
2
votes
1answer
290 views

Network analysis: Formal definition of the number of 1st order neighbours of degree 1

I am looking for a formal definition of a network metric I am using in a scientific article. Let $i$ be a vertex in a graph $G$ and $N(i)$ are the first order neighbors of vertex $i$. I am interested ...

1
2 3 4 5