Skip to main content
Share Your Experience: Take the 2024 Developer Survey
28 votes

What are graph embedding?

What are graph Embeddings ? "Graph Embeddings" is a hot area today in machine learning. It basically means finding "latent vector representation" of graphs which captures the topology (in very basic ...
mausamsion's user avatar
  • 1,282
22 votes
Accepted

What are graph embedding?

Graph embedding learns a mapping from a network to a vector space, while preserving relevant network properties. Vector spaces are more amenable to data science than graphs. Graphs contain edges and ...
Brian Spiering's user avatar
17 votes

Anconda R version - How to upgrade to 4.0 and later

You need to create a new environment and then you can install R 4.+ in Anaconda. Follow these steps. conda create --name r4-base After activating ...
Rheatey Bash's user avatar
13 votes
Accepted

In elbow curve how to find the point from where the curve starts to rise?

TL;DR Use the two functions from below to get the index of the elbow: elbow_index = find_elbow(data, get_data_radiant(data)) **Edit:** I put all of the code below ...
georg-dev's user avatar
  • 1,231
10 votes

How to use Scikit-Learn Label Propagation on graph structured data?

Answering my own question here, as I hope it will be useful to some readers. Scikit-learn is primarily designed to deal with vector structured data. Hence, if you want to perform label propagation/...
Thibaud M's user avatar
  • 101
9 votes
Accepted

Large Graphs: NetworkX distributed alternative

Good , old and unsolved question! Distributed processing of large graphs as far as I know (speaking as a graph guy) has 2 different approaches, with the knowledge of Big Data frameworks or without it. ...
Kasra Manshaei's user avatar
6 votes

Large Graphs: NetworkX distributed alternative

In this present moment, Apache has develop a powerfull API called PySpark. And you can setup Graphframes directly from pyspark command line. Launch from you shell ...
Emanuel Fontelles's user avatar
6 votes
Accepted

Visualize graph network with more than 30k edges

Yes There are! Networkx I think 20k-30k node-edge would be OK on Networkx, IF YOU HAVE A GOOD MACHINE! Networkx is a great library in Python particularly for Graph Analysis so you have access to ...
Kasra Manshaei's user avatar
5 votes

Finding groups of friends in social network data

Community Detection and Clique Percolation: This is a community detection problem. Here is a very detailed review article surveying the state of the art. The Clique Percolation Method is also ...
Syed Ali Hamza's user avatar
5 votes
Accepted

Use cases for graph algorithms and graph data structures in finance and banking

There are many use cases of graph theory in Finance industry and it is a very broad question. As Emre said can be used for Fraud Detection, Risk Modelling, Economic Networks etc. These below links ...
Toros91's user avatar
  • 2,392
4 votes
Accepted

Organize TSNE data into grid

There seem to be a few options, but I found rasterfairy which is very easy to install and use. Has the added bonus of being able to fit to a rectangular grid, but also circular and other arbitrary ...
JeffThompson's user avatar
4 votes
Accepted

Clustering Multiple Networks

Your question is not clear in a way there are two different Graph Clustering problems. One is having a dataset of different graphs and you would like to cluster similar graphs (in this case each ...
Kasra Manshaei's user avatar
4 votes

Unable to generate error bars with seaborn

Try this: g = sns.barplot(y="algorithm", x="avg_weightedcost", hue="group", xerr=df[std_weight]*1, capsize=.2, data=df) Hope it helps
Sagar Dawda's user avatar
4 votes
Accepted

Struggling to understand GCNNs (Graph Convolutional Neural Networks)

The name "Graph Convolutional Neural Network" is a bit misleading, as no "traditional" convolutions (like in the context of CNNs) take place at all. You are correct that it doesn't really make sense ...
timleathart's user avatar
  • 3,940
3 votes
Accepted

How do you set sigma for the Gaussian similarity kernel?

Updated Answer According to a reference paper in Spectral Clustering (von Luxburg) the $\sigma$ is simply set to 1. A further tuning can be applied with some visualization inspection but I did not ...
Kasra Manshaei's user avatar
3 votes

Graph to display differences (or lack of) in multilevel categorical data

OK, here are my attempts with R & ggplot2 1 Simple stacked histogram 2 Dodged stacked histogram ~ bacteria 3 Dodged histogram ~ Culture 4 Dodged histogram ~ change 5 Grouped by number of ...
aivanov's user avatar
  • 1,520
3 votes

Unable to generate error bars with seaborn

I found this question when trying to find a way to add my own precalculated, custom error bars (standard deviations) to a Seaborn barplot with grouped values. I finally managed to find a workaround... ...
iTamarit's user avatar
3 votes
Accepted

How to add numbers to the axes of a graph?

Use xticks. e.g. x=np.arange(1,16) y = -60000*(3+np.log(1/x)) plt.plot(x,y,'b') plt.xticks(x) plt.show()
user12075's user avatar
  • 2,284
3 votes
Accepted

Are there algorithms for clustering objects with pairwise distances, without computing all pairwise distances?

Well, one may argue that DBSCAN is based on all pairwise distances, but it uses data indexing to avoid computing all of them using geometric bounds. And there are other examples if you browse through ...
Has QUIT--Anony-Mousse's user avatar
3 votes

Using deep learning on graphs

One of the most popular ways of doing classification on graphs is through graph convolutional networks. By running a convolution over the nodes of a graph, the neural net is able to learn the local ...
Victor Ng's user avatar
  • 370
3 votes

Is there a metric for "cliquiness" for social graphs?

The formal definition of a clique is a fully connected subgraph (or cliquiness=1.0 in your example) where the shortest path between all the nodes in the clique is 1. To relax that, you have use the n-...
Boyd's user avatar
  • 41
2 votes

Partitioning Weighted Undirected Graph

If you are using python, and have created a weighted graph using NetworkX, then you can use python-louvain for clustering. Where G is a weighted graph: ...
kingledion's user avatar
2 votes
Accepted

Libraries for (label propagation algorithms/frequent subgraph mining) for graphs in R

This is an old post, but there is a subgraph package and accompanying book/documentation for doing this in R: https://www.csc.ncsu.edu/faculty/samatova/practical-graph-mining-with-R/...
jojo's user avatar
  • 36
2 votes

Visualizing a graph with a million vertices

I would recommend to use Graphexp. Gephi is highly dependent on the RAM of your computer which is obviously limited. Graphexp on the other hand display's only a limited number of Nodes, through which ...
Sandeep Choudhary's user avatar
2 votes

How to approach graphed data for a binary classification system?

Super comprehensive question. You are basically asking for tones of directions! I'd suggest to start with link prediction problem. So assume you have a directed/signed/weighted graph and you want to ...
Kasra Manshaei's user avatar
2 votes

Which algorithms should I use for recommendation system using a graph database?

A biadjacency matrix of a bipartite graph admits matrix factorisation. For $m$ items and $n$ users, the biadjacency matrix is an $m \times n$ matrix which can be factorised into two lower-rank ...
R Hill's user avatar
  • 1,105
2 votes

Why do we do citation analysis?

One simple example of an application of citation analysis is ranking. Imagine you do a keyword search over a collection of scientific articles, How do you rank the result set from most relevant to ...
Dani Mesejo's user avatar
  • 2,226
2 votes

In a directed graph, how to measure whether a node is more "upstream" or "downstream"?

Please clarify what you are looking for in the presence of cycles. I'd assume the cycles aren't standalone- any node in a cycle could also have inputs from further upstream, and could also send ...
gms's user avatar
  • 181
2 votes
Accepted

Labeling hubs in a network

Im not sure if I understand your question. But, you do not need to create a graph to work out node degree. Node degree can be obtained by taking the sum of the row or column in the adjacency matrix. E....
BenP's user avatar
  • 317
2 votes

Visualizing a large graph (10'000 nodes)

Gephi may be able to handle it. Graphistry definitely can but you may need to pay. Ultimately though, unless it fits some specific graph structures, most if not all major algorithms will produce an ...
Gabe's user avatar
  • 21

Only top scored, non community-wiki answers of a minimum length are eligible