| Title: | Weighted, Two-Mode, and Longitudinal Networks Analysis |
|---|---|
| Description: | Binary ties limit the richness of network analyses as relations are unique. The two-mode structure contains a number of features lost when projection it to a one-mode network. Longitudinal datasets allow for an understanding of the causal relationship among ties, which is not the case in cross-sectional datasets as ties are dependent upon each other. |
| Authors: | Tore Opsahl |
| Maintainer: | Tore Opsahl <[email protected]> |
| License: | GPL-3 |
| Version: | 3.0.16 |
| Built: | 2025-02-13 04:42:32 UTC |
| Source: | https://github.com/cran/tnet |
This dataset contains two intra-organizational networks from a consulting company (46 employees). These networks was used by Cross and Parker (2004).
In the first network, the ties are differentiated on a scale from 0 to 5 in terms of frequency of information or advice requests ("Please indicate how often you have turned to this person for information or advice on work-related topics in the past three months"). 0: I Do Not Know This Person; 1: Never; 2: Seldom; 3: Sometimes; 4: Often; and 5:Very Often.
In the second network, ties are differentiated in terms of the value placed on the information or advice received ("For each person in the list below, please show how strongly you agree or disagree with the following statement: In general, this person has expertise in areas that are important in the kind of work I do."). The weights in this network is also based on a scale from 0 to 5. 0: I Do Not Know This Person; 1: Strongly Disagree; 2: Disagree; 3: Neutral; 4: Agree; and 5: Strongly Agree.
In addition to the relational data, the dataset also contains information about the people (nodal attributes). The following attributes are known: the organisational level (1 Research Assistant; 2: Junior Consultant; 3: Senior Consultant; 4: Managing Consultant; 5: Partner), gender (1: male; 2: female), region (1: Europe; 2: USA), and location (1: Boston; 2: London; 3: Paris; 4: Rome; 5: Madrid; 6: Oslo; 7: Copenhagen).
See http://toreopsahl.com/datasets/
Cross.Parker.Consulting.net.info Cross.Parker.Consulting.net.value Cross.Parker.Consulting.node.gender Cross.Parker.Consulting.node.location Cross.Parker.Consulting.node.orglevel Cross.Parker.Consulting.node.regionCross.Parker.Consulting.net.info Cross.Parker.Consulting.net.value Cross.Parker.Consulting.node.gender Cross.Parker.Consulting.node.location Cross.Parker.Consulting.node.orglevel Cross.Parker.Consulting.node.region
The networks are data frames with three columns. The first column is the id of the sender, the second column is the id of the receiver, and the third column is the weight of the tie. The nodal attributes are vectors.
Cross, R., Parker, A., 2004. The Hidden Power of Social Networks. Harvard Business School Press, Boston, MA.
http://toreopsahl.com/datasets/
This network is the Facebook-like Social Network-dataset used in my Ph.D. thesis. This network has also been described in Patterns and Dynamics of Users' Behaviour and Interaction: Network Analysis of an Online Community and used in Prominence and control: The weighted rich-club effect and Clustering in weighted networks. The network originates from a virtual community among students at University of California, Irvine. The edgelists include the users that sent or received at least one message during that period (1,899). A total number of 59,835 online messages were sent among over 20,296 directed ties.
OnlineSocialNetwork.n1899.net OnlineSocialNetwork.n1899.lnetOnlineSocialNetwork.n1899.net OnlineSocialNetwork.n1899.lnet
OnlineSocialNetwork.n1899.net: A data frame with three columns. The first column is the id of the sender, the second column is the id of the receiver, and the third column is the weight of the tie.
OnlineSocialNetwork.n1899.lnet: A data frame with four columns. The first column is the timestamp, the second column is the id of the sender, the third column is the id of the receiver, and the fourth column is the weight of the tie (always 1).
Tore Opsahl; http://toreopsahl.com
Opsahl, T., Panzarasa, P., 2009. Clustering in weighted networks. Social Networks 31 (2), 155-163, doi: 10.1016/j.socnet.2009.02.002
http://toreopsahl.com/datasets/
This package is created to analyse weighted networks, two-mode networks, and longitudinal networks datasets. Binary ties limit the richness of network analyses as relations are unique. The two-mode structure contains a number of features lost when projection it to a one-mode network. Longitudinal datasets allow for an understanding of the causal relationship among ties, which is not the case in cross-sectional datasets as ties are dependent upon each other.
| Package: | tnet |
| Type: | Package |
| Version: | 3.0.16 |
| Date: | 2020-02-23 |
This package is created to analyse weighted networks, two-mode networks, and longitudinal networks datasets. More information is available on http://toreopsahl.com/tnet/
It utilises three forms of data structures (it can automatically convert matrices etc into these formats, see the as.tnet-function):
1) simple weighted data in the following format (creator.node.id target.node.id tie.weight):
1 2 4
1 3 2
Note: For undirected networks, each tie must be mentioned twice (see the symmetrise_w-function). For example,
1 2 4
2 1 4
1 3 2
3 1 2
2) two-mode data in the following format (primary.node.id secondar.node.id tie.weight.optional):
1 1 1
2 1 2
3) timed data in the following format (MySQL-timestamp.as.character.string creator.node.id target.node.id tie.weight):
"2007-09-12 13:45:00" 1 2 1
"2007-09-12 13:46:31" 1 2 1
If ties are repeated, the tie increases the weighted. The weight column decides how much weight is added at each time (this can take a negative value to decrease the weight).
Attribute files are read as follows:
0 1 3
0 3 2
1 3 3
where the first row refers to node 1, the second row to node 2, etc. The first column refers to the first attribute, second column to the second attribute and so on.
A big thank you to the igraph guys as this package relies on their work for many of the more computational tasks!
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/tnet/
# Generate a random weighted graph rg <- rg_w(nodes=100,arcs=300,directed=TRUE) # Calculate clustering coefficient clustering_w(rg)# Generate a random weighted graph rg <- rg_w(nodes=100,arcs=300,directed=TRUE) # Calculate clustering coefficient clustering_w(rg)
This function adds negative ties (i.e., a smoothing window) to a longitudinal network.
add_window_l(net,window=21, remove.nodes=TRUE)add_window_l(net,window=21, remove.nodes=TRUE)
net |
Longitudinal network |
window |
Number of days before ties 'expire'. |
remove.nodes |
Whether or not nodes should be removed from the network if they have no more ties. This function adds a self-loop with a negative weight at the time of a node's last tie plus the length of the window. |
.
Returns the longitudinal network with negative arcs.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
t <- c('2007-09-12 13:45:00', '2007-09-12 13:46:31', '2007-09-12 13:47:54', '2007-09-12 13:48:21', '2007-09-12 13:49:27', '2007-09-12 13:58:14', '2007-09-12 13:52:17', '2007-09-12 13:56:59'); i <- c(1,1,1,1,1,1,1,1); j <- c(2,2,2,2,2,2,3,3); w <- c(1,1,1,1,1,1,1,1); sample <- data.frame(t, i, j, w); ## Run the programme add_window_l(sample, window=21)t <- c('2007-09-12 13:45:00', '2007-09-12 13:46:31', '2007-09-12 13:47:54', '2007-09-12 13:48:21', '2007-09-12 13:49:27', '2007-09-12 13:58:14', '2007-09-12 13:52:17', '2007-09-12 13:56:59'); i <- c(1,1,1,1,1,1,1,1); j <- c(2,2,2,2,2,2,3,3); w <- c(1,1,1,1,1,1,1,1); sample <- data.frame(t, i, j, w); ## Run the programme add_window_l(sample, window=21)
This function transforms a longitudinal network to a static edgelist
as.static.tnet(ld)as.static.tnet(ld)
ld |
Longitudinal network |
Returns the data in an edgelist format.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
t <- c('2007-09-12 13:45:00', '2007-09-12 13:46:31', '2007-09-12 13:47:54', '2007-09-12 13:48:21', '2007-09-12 13:49:27', '2007-09-12 13:58:14', '2007-09-12 13:52:17', '2007-09-12 13:56:59'); i <- c(1,1,1,1,1,1,1,1); j <- c(2,2,2,2,2,2,3,3); w <- c(1,1,1,1,1,-1,1,1); net <- data.frame(t, i, j, w); ## Run the programme as.static.tnet(net)t <- c('2007-09-12 13:45:00', '2007-09-12 13:46:31', '2007-09-12 13:47:54', '2007-09-12 13:48:21', '2007-09-12 13:49:27', '2007-09-12 13:58:14', '2007-09-12 13:52:17', '2007-09-12 13:56:59'); i <- c(1,1,1,1,1,1,1,1); j <- c(2,2,2,2,2,2,3,3); w <- c(1,1,1,1,1,-1,1,1); net <- data.frame(t, i, j, w); ## Run the programme as.static.tnet(net)
Checks that a network conforms to the tnet stardards, and attaches a label. If the type parameter is not set, the network is assumed to be a binary two-mode network, a weighted one-mode network, or a longitudinal network if there are 2, 3, or 4 columns respectively. Moreover, if a matrix is entered (more than 4 columns and rows), it is assumed to be a weighted one-mode network if square or a two-mode network if non-square.
as.tnet(net, type=NULL)as.tnet(net, type=NULL)
net |
A network in an edgelist or matrix format. It can be a weighted one-mode network, a binary two-mode network, a weighted two-mode netork, or a longitudinal network. If the data-object has two-columns, it is assumed to be a binary two-mode network; three columns, weighted one-mode network; four columns, longitudinal; five or more and the same number of rows and columns, weighted one-mode network; five or more and –not– the same number of rows and columns, it is assumed to be a two-mode network. |
type |
If you would like to specify the type of network. This could be "weighted one-mode tnet", "binary two-mode tnet", "weighted two-mode tnet", or "longitudinal tnet". |
Returns the network with an attached lable.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
## Load sample data sample <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme as.tnet(sample)## Load sample data sample <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme as.tnet(sample)
This function calculates betweenness scores for nodes in a weighted network based on the distance_w-function.
Note: This algorithm relies on the igraphs package's implementation of Dijkstra's algorithm. Currently, it does not find multiple shortest paths if two exist.
betweenness_w(net, directed=NULL, alpha=1)betweenness_w(net, directed=NULL, alpha=1)
net |
A weighted edgelist |
directed |
logical, whether the network is directed or undirected. Default is NULL, this means that the function checks whether the edgelist is directed or not. |
alpha |
sets the alpha parameter in the generalised measures from Opsahl, T., Agneessens, F., Skvoretz, J., 2010. Node Centrality in Weighted Networks: Generalizing Degree and Shortest Paths. Social Networks. If this parameter is set to 1 (default), the Dijkstra shortest paths are used. The length of these paths rely simply on the tie weights and disregards the number of nodes on the paths. |
Returns a data.frame with two columns: the first column contains the nodes' ids, and the second column contains the nodes' betweenness scores.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2009/02/20/betweenness-in-weighted-networks/
## Load sample data sampledata <- rbind( c(1,2,1), c(1,3,5), c(2,1,1), c(2,4,6), c(3,1,5), c(3,4,10), c(4,2,6), c(4,3,10)) ## Run the programme betweenness_w(sampledata)## Load sample data sampledata <- rbind( c(1,2,1), c(1,3,5), c(2,1,1), c(2,4,6), c(3,1,5), c(3,4,10), c(4,2,6), c(4,3,10)) ## Run the programme betweenness_w(sampledata)
This dataset contains the neural network of the Caenorhabditis elegans worm (C.elegans). It was studied by Watts and Strogatz (1998). The network contains 306 nodes that represent neurons. Two neurons are connected if at least one synapse or gap junction exist between them. The weight is the number of synapses and gap junctions. This network was obtained from the Collective Dynamics Group's website.
celegans.n306.netcelegans.n306.net
A data frame with three columns. The first is the id of the sender; the second is the id of the receiver; and the third is the weight of the tie.
Tore Opsahl; http://toreopsahl.com
Watts, D. J., Strogatz, S. H., 1998. Collective dynamics of "small-world" networks. Nature 393, 440-442.
http://toreopsahl.com/datasets/
This function calculates closeness scores for nodes in a weighted network based on the distance_w-function.
closeness_w(net, directed=NULL, gconly=TRUE, precomp.dist=NULL, alpha=1)closeness_w(net, directed=NULL, gconly=TRUE, precomp.dist=NULL, alpha=1)
net |
A weighted edgelist |
directed |
Logical: whether the edgelist is directed or undirected. Default is NULL, then the function detects this parameter. |
gconly |
Logical: whether to calculate closeness only on the main component (traditional closeness). Default is TRUE. If this parameter is set to FALSE, a closeness measure for all nodes is computed. For details, see http://toreopsahl.com/2010/03/20/closeness-centrality-in-networks-with-disconnected-components/ |
precomp.dist |
If you have already computed the distance matrix using distance_w-function, you can enter the name of the matrix-object here. |
alpha |
sets the alpha parameter in the generalised measures from Opsahl, T., Agneessens, F., Skvoretz, J., 2010. Node Centrality in Weighted Networks: Generalizing Degree and Shortest Paths. Social Networks. If this parameter is set to 1 (default), the Dijkstra shortest paths are used. The identification procedure of these paths rely simply on the tie weights and disregards the number of nodes on the paths. |
Returns a data.frame with three columns: the first column contains the nodes' ids, the second column contains the closeness scores, and the third column contains the normalised closeness scores (i.e., divided by N-1).
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2009/01/09/average-shortest-distance-in-weighted-networks/
## Load sample data sampledata <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme closeness_w(sampledata)## Load sample data sampledata <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme closeness_w(sampledata)
This function calculates the local two-mode clusering coefficient as proposed in Opsahl, T., 2010. Triadic closure in two-mode networks: Redefining the global and local clustering coefficients. arXiv:1006.0887.
clustering_local_tm(net)clustering_local_tm(net)
net |
A binary or weighted two-mode edgelist |
Returns the local clustering coefficient for the primary node set (the first of an edgelist or the rows of a matrix)
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Opsahl, T., 2010. Triadic closure in two-mode networks: Redefining the global and local clustering coefficients. arXiv:1006.0887
# Weighted two-mode network net <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) ## Run binary clustering function clustering_local_tm(net[,1:2]) ## Run weighted clustering function clustering_local_tm(net)# Weighted two-mode network net <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) ## Run binary clustering function clustering_local_tm(net[,1:2]) ## Run weighted clustering function clustering_local_tm(net)
This function calculates Barrat et al. (2004) generalised local clusering coefficient.
See http://toreopsahl.com/2009/01/23/weighted-local-clustering-coefficient/ for a detailed description. By default it defines the triplet value as the average of the two tie weights; however it can also define it differently. See the blog post.
Note: If there are very large tie weights in a network, the geometric method in R fails. However, this can be fixed by transforming the values.
net[,"w"] <- (net[,"w"]/min(net[,"w"]))
This step is not required unless you receive warnings when running the function.
clustering_local_w(net, measure = "am")clustering_local_w(net, measure = "am")
net |
A weighted edgelist |
measure |
The measure-switch control the method used to calculate the value of the triplets. |
Returns a data.frame with at least two columns: the first column contains the nodes' ids, and the remaining columns contain the corresponding clustering scores.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Barrat, A., Barthelemy, M., Pastor-Satorras, R., Vespignani, A., 2004. The architecture of complex weighted networks. Proceedings of the National Academy of Sciences 101 (11), 3747-3752. arXiv:cond-mat/0311416
http://toreopsahl.com/2009/01/23/weighted-local-clustering-coefficient/
## Generate a random graph #density: 300/(100*99)=0.03030303; #this should be average from random samples rg <- rg_w(nodes=100,arcs=300,weights=1:10,directed=FALSE) ## Run clustering function clustering_local_w(rg)## Generate a random graph #density: 300/(100*99)=0.03030303; #this should be average from random samples rg <- rg_w(nodes=100,arcs=300,weights=1:10,directed=FALSE) ## Run clustering function clustering_local_w(rg)
This function calculates the two-mode clusering coefficient as proposed by Opsahl, T., 2010. Triadic closure in two-mode networks: Redefining the global and local clustering coefficients. arXiv:1006.0887.
Note: If you are having problems with this function (i.e., run out of memory or it being slow for simulations), there is a quicker and much more memory efficient c++ function. However, this function is not fully integrated in R, and requires a few extra steps. Send me an email to get the source-code and Windows-compiled files.
clustering_tm(net, subsample=1, seed=NULL)clustering_tm(net, subsample=1, seed=NULL)
net |
A binary or weighted two-mode edgelist |
subsample |
Whether a only a subset of 4-paths should we used when calculating the measure. This is particularly useful when running out of memory analysing large networks. If it is set to 1, all the 4-paths are analysed. If it set to a value below one, this is roughly the proportion of 4-paths that will be analysed. If it is set to an interger greater than 1, this number of ties that form the first part of a 4-path that will be analysed. Note: The c++ functions are better as they analyse the full network. |
seed |
If a subset of 4-paths is analysed, by setting this parameter, the results are reproducable. |
Returns the outcome of the equation presented in the paper
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Tore Opsahl. Triadic closure in two-mode networks: Redefining the global and local clustering coefficients. arXiv:1006.0887
# Weighted two-mode network net <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) ## Run binary clustering function clustering_tm(net[,1:2]) ## Run weighted clustering function clustering_tm(net)# Weighted two-mode network net <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) ## Run binary clustering function clustering_tm(net[,1:2]) ## Run weighted clustering function clustering_tm(net)
This function calculates the generalised clusering coefficient as proposed by Opsahl, T., Panzarasa, P., 2009. Clustering in weighted networks. Social Networks 31 (2), 155-163, doi: 10.1016/j.socnet.2009.02.002
Note: If you are having problems with this function (i.e., run out of memory or it being slow for simulations), there is a quicker and much more memory efficient c++ function. However, this function is not fully integrated in R, and requires a few extra steps. Send me an email to get the source-code and Windows-compiled files.
clustering_w(net, measure = "am")clustering_w(net, measure = "am")
net |
A weighted edgelist |
measure |
The measure-switch control the method used to calculate the value of the triplets. |
Returns the outcome of the equation presented in the paper for the method specific (measure)
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Opsahl, T., Panzarasa, P., 2009. Clustering in weighted networks. Social Networks 31 (2), 155-163, doi: 10.1016/j.socnet.2009.02.002
http://toreopsahl.com/2009/04/03/article-clustering-in-weighted-networks/
## Generate a random graph #density: 300/(100*99)=0.03030303; #this should be average from random samples rg <- rg_w(nodes=100,arcs=300,weights=1:10) ## Run clustering function clustering_w(rg)## Generate a random graph #density: 300/(100*99)=0.03030303; #this should be average from random samples rg <- rg_w(nodes=100,arcs=300,weights=1:10) ## Run clustering function clustering_w(rg)
The compress_ids function removes non-active nodes from one-mode/two-mode/longitudinal networks.
compress_ids(net,type=NULL)compress_ids(net,type=NULL)
net |
A network in an edgelist or matrix format. See as.tnet |
type |
See as.tnet |
Returns a list with either 2 or 3 objects. The first one is the network with the compressed id. The second object is the translation table between the original node identification numbers and the newly assigned. For two-mode networks, the second object is the translation table of the primary nodes, and the third object is the translation table for the secondary nodes .
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/
## Load sample data t <- c("2007-09-12 13:45:00", "2007-09-12 13:45:00", "2007-09-12 13:45:01", "2007-09-12 13:46:31", "2007-09-12 13:46:31", "2007-09-12 13:47:54", "2007-09-12 13:48:21", "2007-09-12 13:49:27", "2007-09-12 13:49:27", "2007-09-12 13:52:17", "2007-09-12 13:56:59", "2007-09-12 13:58:14") i <- c(1,2,1,3,1,2,3,5,1,3,1,1); j <- c(1,2,2,3,3,1,2,5,5,2,3,5); w <- c(1,1,1,1,1,1,1,1,1,1,1,1); samplenet <- data.frame(t, i, j, w); ## Run the function compress_ids(samplenet)## Load sample data t <- c("2007-09-12 13:45:00", "2007-09-12 13:45:00", "2007-09-12 13:45:01", "2007-09-12 13:46:31", "2007-09-12 13:46:31", "2007-09-12 13:47:54", "2007-09-12 13:48:21", "2007-09-12 13:49:27", "2007-09-12 13:49:27", "2007-09-12 13:52:17", "2007-09-12 13:56:59", "2007-09-12 13:58:14") i <- c(1,2,1,3,1,2,3,5,1,3,1,1); j <- c(1,2,2,3,3,1,2,5,5,2,3,5); w <- c(1,1,1,1,1,1,1,1,1,1,1,1); samplenet <- data.frame(t, i, j, w); ## Run the function compress_ids(samplenet)
This dataset contains two intra-organizational networks from a research team in a manufacturing company (77 employees). These networks was used by Cross and Parker (2004).
In the first network, the ties among the researchers are differentiated in terms of advice ("Please indicate the extent to which the people listed below provide you with information you use to accomplish your work"). The weights are based on the following scale: 0: I Do Not Know This Person/I Have Never Met this Person; 1: Very Infrequently; 2: Infrequently; 3: Somewhat Infrequently; 4: Somewhat Frequently; 5: Frequently; and 6: Very Frequently.
The second network is based on the employees' awareness of each others' knowledge and skills ("I understand this person's knowledge and skills. This does not necessarily mean that I have these skills or am knowledgeable in these domains but that I understand what skills this person has and domains they are knowledgeable in"). The weight scale in this network is: 0: I Do Not Know This Person/I Have Never Met this Person; 1: Strongly Disagree; 2: Disagree; 3: Somewhat Disagree; 4: Somewhat Agree; 5: Agree; and 6: Strongly Agree.
In addition to the relational data, the dataset also contains information about the people (nodal attributes). The following attributes are known: location (1: Paris; 2: Frankfurt; 3: Warsaw; 4: Geneva), tenure (1: 1-12 months; 2: 13-36 months; 3: 37-60 months; 4: 61+ months) and the organisational level (1: Global Dept Manager; 2: Local Dept Manager; 3: Project Leader; 4: Researcher).
See http://toreopsahl.com/datasets/
Cross.Parker.Manufacturing.net.info Cross.Parker.Manufacturing.net.aware Cross.Parker.Manufacturing.node.location Cross.Parker.Manufacturing.node.orglevel Cross.Parker.Manufacturing.node.tenureCross.Parker.Manufacturing.net.info Cross.Parker.Manufacturing.net.aware Cross.Parker.Manufacturing.node.location Cross.Parker.Manufacturing.node.orglevel Cross.Parker.Manufacturing.node.tenure
The networks are data frames with three columns. The first column is the id of the sender; the second column is the id of the receiver; and the third column is the weight of the tie. The nodal attributes are vectors.
Cross, R., Parker, A., 2004. The Hidden Power of Social Networks. Harvard Business School Press, Boston, MA.
http://toreopsahl.com/datasets/
This dataset was collected by Davis and colleague in the 1930s. It contains the observed attendance by 18 Southern women (primary nodes) at 14 social events (secondary nodes). This has been projected onto a co-occurance one-mode network, and a one-mode network based on Newman's (2001) method.
Davis.Southern.women.2mode Davis.Southern.women.1mode.Cooccurance Davis.Southern.women.1mode.NewmanDavis.Southern.women.2mode Davis.Southern.women.1mode.Cooccurance Davis.Southern.women.1mode.Newman
The two-mode network is a data frame with two columns (primary nodes and secondary nodes, respectively). The one-mode networks are data frames with three columns: the first column is the id of the sender; the second column is the id of the receiver; and the column third is the weight of the tie.
Tore Opsahl; http://toreopsahl.com
Davis, A., Gardner, B. B., Gardner, M. R., 1941. Deep South. University of Chicago Press, Chicago, IL.
http://toreopsahl.com/datasets/
This function calculates two degree measures: the number of contacts that a node is connected to, and the sum of weights on ties originating from a node (strength).
degree_tm(net,measure=c("degree","output"))degree_tm(net,measure=c("degree","output"))
net |
A two-mode network |
measure |
specifies which measures should be calculated |
Returns a data.frame with two or three columns: the first column contains the nodes' ids, and the remaining columns contain the scores of the measures specified in the measure-parameter.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/blog/
## Load sample data network <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) ## Run the programme degree_tm(network)## Load sample data network <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) ## Run the programme degree_tm(network)
This function calculates two degree measures: the number of contacts that a node is connected to, and the sum of weights on ties originating from a node (out-strength). To calculate the reverse (in-degree, in-strength), specify type="in".
degree_w(net,measure=c("degree","output"), type="out", alpha=1)degree_w(net,measure=c("degree","output"), type="out", alpha=1)
net |
A weighted edgelist |
measure |
specifies which measures should be calculated |
type |
shall out- or in-measures be calculated? Default is out. For undirected networks, this setting is irrelevant, but must be specified. |
alpha |
sets the alpha parameter in the generalised measures from Opsahl, T., Agneessens, F., Skvoretz, J., 2010. Node Centrality in Weighted Networks: Generalizing Degree and Shortest Paths. Social Networks. If this parameter is set to 1 (default), the sum of tie weights is used. This measure simply use the tie weights and disregards the number of nodes on the paths. |
Returns a data.frame with two or more columns: the first column contains the nodes' ids, and the remaining columns contain the scores of the measures specified in the measure-parameter.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2008/11/28/network-weighted-network/
## Load sample data network <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme degree_w(network)## Load sample data network <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme degree_w(network)
The dichotomise function creates a binary two-mode network from a weighted edgelist.
dichotomise_tm(net,GT=0)dichotomise_tm(net,GT=0)
net |
A weighted two-mode network |
GT |
the cut-off parameter. Default is set to 0, so edges/arcs with a weight greater than 0 is set to 1. |
Returns the edgelist with edges below the cut-off removed, and all weights equal to 1.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2008/11/28/network-weighted-network/
## Load sample data sample <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) ## Run the programme dichotomise_tm(sample, GT=2)## Load sample data sample <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) ## Run the programme dichotomise_tm(sample, GT=2)
The dichotomise function creates a binary one-mode network from a weighted edgelist.
dichotomise_w(net,GT=0)dichotomise_w(net,GT=0)
net |
A weighted one-mode network |
GT |
the cut-off parameter. Default is set to 0, so edges/arcs with a weight greater than 0 is set to 1. |
Returns the edgelist with edges below the cut-off removed, and all weights equal to 1.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2008/11/28/network-weighted-network/
## Load sample data sample <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme dichotomise_w(sample, GT=2)## Load sample data sample <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme dichotomise_w(sample, GT=2)
The shortest path length, or geodesic distance, between two nodes in a binary network is the minimum number of steps you need to make to go from one of them to the other. See the distance_w-function for more details.
distance_tm(net, projection.method="sum", gconly=TRUE,subsample=1, seed=NULL)distance_tm(net, projection.method="sum", gconly=TRUE,subsample=1, seed=NULL)
net |
A two-mode network |
projection.method |
The way the two-mode network is projected. The sum method defines tie weights as the number of common nodes (e.g., events, projects etc) that two individuals had contact through. In certain cases, "Newman" might be better. See the projecting_tm-function. |
gconly |
logical, whether the function should only be calculated for the giant component. Default is TRUE. |
subsample |
Whether a only a subset of starting nodes should we used when calculating the measure. This is particularly useful when running out of memory analysing large networks. If it is set to 1, all distances are analysed. If it set to a value below one, this is roughly the proportion of starting noes that will be analysed. If it is set to an interger greater than 1, this number of starting nodes that will be analysed. |
seed |
If a subset of starting nodes is analysed, by setting this parameter, the results are reproducable. |
Returns a distance matrix.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2009/01/09/average-shortest-distance-in-weighted-networks/
# Load networks net <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) # Run the function distance_tm(net)# Load networks net <- cbind( i=c(1,1,2,2,2,3,3,4,5,5,6), p=c(1,2,1,3,4,2,3,4,3,5,5), w=c(3,5,6,1,2,6,2,1,3,1,2)) # Run the function distance_tm(net)
The shortest path length, or geodesic distance, between two nodes in a binary network is the minimum number of steps you need to make to go from one of them to the other. This distance is the quickest connection between nodes when all ties are the same. However, in a weighted network, all ties are not the same. See http://toreopsahl.com/2009/01/09/average-shortest-distance-in-weighted-networks/ for more deatails.
distance_w(net, directed=NULL, gconly=TRUE, subsample=1, seed=NULL)distance_w(net, directed=NULL, gconly=TRUE, subsample=1, seed=NULL)
net |
A weighted edgelist |
directed |
logical, whether the network is directed or undirected. Default is NULL, this means that the function checks whether the edgelist is directed or not. |
gconly |
logical, whether the function should only be calculated for the giant component. Default is TRUE. |
subsample |
Whether a only a subset of starting nodes should we used when calculating the measure. This is particularly useful when running out of memory analysing large networks. If it is set to 1, all distances are analysed. If it set to a value below one, this is roughly the proportion of starting noes that will be analysed. If it is set to an interger greater than 1, this number of starting nodes that will be analysed. |
seed |
If a subset of starting nodes is analysed, by setting this parameter, the results are reproducable. |
Returns a distance matrix.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2009/01/09/average-shortest-distance-in-weighted-networks/
## Load sample data sample <- rbind( c(1,2,8), c(1,4,1), c(2,1,8), c(2,3,6), c(3,2,6), c(3,4,10), c(4,1,1), c(4,3,10)) ## Run the programme distance_w(sample)## Load sample data sample <- rbind( c(1,2,8), c(1,4,1), c(2,1,8), c(2,3,6), c(3,2,6), c(3,4,10), c(4,1,1), c(4,3,10)) ## Run the programme distance_w(sample)
Freeman's EIES networks (Freeman, 1979) was the main network used in Wasserman and Faust (1994). This dataset was collected in 1978 and contains three networks of researchers working on social network analysis. The first network contains the personal relationships among 48 of the researchers at the beginning of the study (time 1). The second network is the personal relationship at the end of the study (time 2). In these two networks, all ties have a weight between 0 and 4. 4 represents a close personal friend of the researcher's; 3 represents a friend; 2 represents a person the researcher has met; 1 represents a person the researcher has heard of, but not met; and 0 represents a person unknown to the researcher. The third network is different. It is a matrix with the number of messages sent among 32 of the researchers that used an electronic communication tool (frequency matrix).
There are two pieces of information about each of the 32 researchers that were part of the third network (nodal attributes): the main disciplinary affiliation (1: sociology; 2: anthropology; 3: mathematics or statistics; and 4: others) and the number of citations each researcher had in the Social Science Citation Index in 1978.
See http://toreopsahl.com/datasets/
Freemans.EIES.net.1.n48 Freemans.EIES.net.2.n48 Freemans.EIES.net.3.n32 Freemans.EIES.node.Name.n32 Freemans.EIES.node.Citations.n32 Freemans.EIES.node.Discipline.n32Freemans.EIES.net.1.n48 Freemans.EIES.net.2.n48 Freemans.EIES.net.3.n32 Freemans.EIES.node.Name.n32 Freemans.EIES.node.Citations.n32 Freemans.EIES.node.Discipline.n32
The networks are data frames with three columns. The first column is the id of the sender, the second column is the id of the receiver; and the third column is the weight of the tie. The attributes are vectors.
Freeman, S.C., Freeman, L.C., 1979. The networkers network: A study of the impact of a new communications medium on sociometric structure. Social Science Research Reports 46. University of California, Irvine, CA.
See http://toreopsahl.com/datasets/
This function identifies growth mechanisms responsible for tie generation in longitudinal networks.
growth_l(net, perspective = "actor", effects, window=NULL, binary=FALSE, nstrata=10, seed=NULL, regression=TRUE)growth_l(net, perspective = "actor", effects, window=NULL, binary=FALSE, nstrata=10, seed=NULL, regression=TRUE)
net |
A longitudinal network |
perspective |
whether an actor or dyadic perspective should be used (i.e., whether the network is directed or undirected). Currently, only the actor perspective is included. |
effects |
The effects to be analysed |
window |
Whether a window should be used. |
binary |
Whether duplicated ties should be removed. |
nstrata |
Total number of regression observations for each observed tie (i.e., number of control cases plus 1 for the observed tie). Minimum is 2 in which 1 control case is used for each observed case. |
seed |
seed for random generator, set to have reproducable results. |
regression |
Whether R should perform the regression or output a regression table. If you want to run multiple regression, it is quicker to output the table, and then run multiple regressions. By outputting the table, it is also possible to add square terms and additional data. |
Returns a regression result or table.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Tore Opsahl, Bernie Hogan. Growth mechanisms in continuously-observed networks: Communication in a Facebook-like community. arXiv:1010.2141
## Load sample data t <- c('2007-09-12 13:45:00', '2007-09-12 13:46:31', '2007-09-12 13:47:54', '2007-09-12 13:48:21', '2007-09-12 13:49:27', '2007-09-12 13:58:14', '2007-09-12 13:52:17', '2007-09-12 13:56:59'); i <- c(1,1,2,3,1,3,1,1); j <- c(2,3,1,2,4,2,3,4); w <- c(1,1,1,1,1,1,1,1); sample <- data.frame(t, i, j, w); ## Run the function growth_l(sample, effects="indegree", nstrata=2)## Load sample data t <- c('2007-09-12 13:45:00', '2007-09-12 13:46:31', '2007-09-12 13:47:54', '2007-09-12 13:48:21', '2007-09-12 13:49:27', '2007-09-12 13:58:14', '2007-09-12 13:52:17', '2007-09-12 13:56:59'); i <- c(1,1,2,3,1,3,1,1); j <- c(2,3,1,2,4,2,3,4); w <- c(1,1,1,1,1,1,1,1); sample <- data.frame(t, i, j, w); ## Run the function growth_l(sample, effects="indegree", nstrata=2)
This is the co-authorship network of scientists based on preprints posted to Condensed Matter section of arXiv E-Print Archive between 1995 and 1999.
This network can be classified as a two-mode or affiliation network since there are two types of "nodes" (authors and papers) and connections exist only between different types of nodes. An author is connected to a paper if her or his name appeared on it.
Few network measures exist for two-mode networks, and therefore, these networks are often projected onto a one-mode (only one type of nodes) network by selecting one of the types of nodes and linking two nodes if they were connected to the same node (of the other kind).
Traditionally, the ties in projected one-mode networks do not have weights. Recent empirical studies of two-mode networks has created a weighted network by defining the weights as the number of co-occurrences (e.g., the number of papers that two authors had collaborated on).
This method was refined by Newman (2001). He argued that smaller collaborations created stronger social bonds among scientists than larger ones. Therefore, he extended this procedure and proposed to define weights among the nodes use the following formula:
w_ij = sum_p 1/(N_p -1)
where w_ij is the weight between node i and node j, p is the papers that they have collaborated on, and N_p is the number of authors on a paper. This implies that if two authors only write a single paper together with no other co-authors, they get a weight of 1. However, if they have a co-author, the weight on the tie between them is 0.5. If two authors have written two papers together without any co-author, the weight of their tie would be 2. A more complicated example is the tie between two authors who have written two papers together: one without any other co-author and one with one co-author. The first paper would give their tie a weight of 1, and the second tie would add 0.5 to the weight of this tie. Therefore, the weight is 1.5.
Note: This method has been explained in more detail in the following post:
http://toreopsahl.com/2009/05/01/projecting-two-mode-networks-onto-weighted-one-mode-networks/
This is the two-mode network.
See http://toreopsahl.com/datasets/
Newman.Condmat.95.99.net.2mode Newman.Condmat.95.99.net.1mode.wNewmanNewman.Condmat.95.99.net.2mode Newman.Condmat.95.99.net.1mode.wNewman
The two-mode network is a data frame with two columns. The first column is the id of authors and the second column is the id of papers. The one-mode network is a data frame with three columns. The first two columns are ids of the authors, and the third column is the weight of the tie. This is calculated based on Newman's (2001) method for defining tie weights. See the projecting_tm-function.
Newman, M. E. J., 2001. The structure of scientific collaboration networks. Proceedings of the National Academy of Sciences of the United States of America 98, 404-409.
See http://toreopsahl.com/datasets/
This function is the implemtation of the procedure outlined on
http://toreopsahl.com/2009/05/01/projecting-two-mode-networks-onto-weighted-one-mode-networks/
projecting_tm(net, method = "sum")projecting_tm(net, method = "sum")
net |
A two-mode edgelist |
method |
The method-switch control the method used to calculate the weights. |
Returns a one-mode network
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Opsahl. T., 2009. Projecting two-mode networks onto weighted one-mode networks. Available at: http://toreopsahl.com/2009/05/01/projecting-two-mode-networks-onto-weighted-one-mode-networks/
## define two-mode network two.mode.net <- cbind( i=c(1,1,2,2,2,2,2,3,4,5,5,5,6), p=c(1,2,1,2,3,4,5,2,3,4,5,6,6)) ## Run the function projecting_tm(two.mode.net, method="Newman")## define two-mode network two.mode.net <- cbind( i=c(1,1,2,2,2,2,2,3,4,5,5,5,6), p=c(1,2,1,2,3,4,5,2,3,4,5,6,6)) ## Run the function projecting_tm(two.mode.net, method="Newman")
The reinforcement_tm-function computes Robin and Alexander's (2004) 4-cycle metric for two-mode networks.
reinforcement_tm(net)reinforcement_tm(net)
net |
A two-mode network |
Returns the score
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/tnet/two-mode-networks/clustering/
## Load sample data net <- rg_tm(10, 7, 0.4) ## Run the programme reinforcement_tm(net)## Load sample data net <- rg_tm(10, 7, 0.4) ## Run the programme reinforcement_tm(net)
This function reshuffles a longitudinal dataset.
rg_reshuffling_l(net, keep.i = FALSE, keep.j = FALSE, seed = NULL)rg_reshuffling_l(net, keep.i = FALSE, keep.j = FALSE, seed = NULL)
net |
Longitudinal network |
keep.i |
Whether or not the tie creators should be maintained |
keep.j |
Whether or not the tie receivers should be maintained |
seed |
the random seed. If you want it to have reproducable result, set using an integer |
Returns a reshuffled longitudinal network
version 1.0.0
Tore Opsahl; http://toreopsahl.com
t <- c('2007-09-12 13:45:00', '2007-09-12 13:46:31', '2007-09-12 13:47:54', '2007-09-12 13:48:21', '2007-09-12 13:49:27', '2007-09-12 13:58:14', '2007-09-12 13:52:17', '2007-09-12 13:56:59'); i <- c(1,1,1,1,1,1,1,1); j <- c(2,2,2,2,2,2,3,3); w <- c(1,1,1,1,1,1,1,1); sample <- data.frame(t, i, j, w); rg_reshuffling_l(sample)t <- c('2007-09-12 13:45:00', '2007-09-12 13:46:31', '2007-09-12 13:47:54', '2007-09-12 13:48:21', '2007-09-12 13:49:27', '2007-09-12 13:58:14', '2007-09-12 13:52:17', '2007-09-12 13:56:59'); i <- c(1,1,1,1,1,1,1,1); j <- c(2,2,2,2,2,2,3,3); w <- c(1,1,1,1,1,1,1,1); sample <- data.frame(t, i, j, w); rg_reshuffling_l(sample)
This function randomly resuffles a binary two-mode edgelist whilst maintaining each nodes' degree (both primary and secondary nodes).
rg_reshuffling_tm(net, option="links", seed=NULL)rg_reshuffling_tm(net, option="links", seed=NULL)
net |
A two-mode network |
option |
Either link reshuffling (option="links") or weight reshuffling (option="weights"), see Opsahl et al. (2008). |
seed |
seed for random generator, set if you want random yet reproducable results. |
Returns a binary two-mode edgelist.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2009/05/29/weighted-rich-club-effect-a-more-appropriate-null-model-for-scientific-collaboration-networks/
## Load data net <- rg_tm(10, 8, 0.4) ## Run the function on a subset rg_reshuffling_tm(net, seed=1)## Load data net <- rg_tm(10, 8, 0.4) ## Run the function on a subset rg_reshuffling_tm(net, seed=1)
This function randomly resuffles a weighted edgelist.
rg_reshuffling_w(net, option="weights", directed=NULL, seed=NULL)rg_reshuffling_w(net, option="weights", directed=NULL, seed=NULL)
net |
A weighted edgelist |
option |
what should be reshuffled: 1) weights (default): randomly assigns the weights to the edges; 2) links: maintain the degree distribution, but changes the contacts randomly. |
directed |
logical: is the network directed or undirected. Default: NULL |
seed |
seed for random generator, set if you want random yet reproducable results. |
Returns a randomised (reshuffled) network.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Molloy, M., Reed, B., 1995. A critical point for random graphs with a given degree sequence. Random Structures and Algorithms 6, 161-180.
Opsahl, T., Colizza, V., Panzarasa, P., Ramasco, J. J., 2008. Prominence and control: The weighted rich-club effect. Physical Review Letters 101 (168702). arXiv:0804.0417.
http://toreopsahl.com/2008/12/12/article-prominence-and-control-the-weighted-rich-club-effect/
## Load sample data sampledata<-rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)); ## Run the function rg_reshuffling_w(sampledata, option="weights", directed=FALSE)## Load sample data sampledata<-rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)); ## Run the function rg_reshuffling_w(sampledata, option="weights", directed=FALSE)
Creates classical random binary and weighted two-mode networks
rg_tm(ni=100,np=100,ties=300,weights=1,seed=NULL)rg_tm(ni=100,np=100,ties=300,weights=1,seed=NULL)
ni |
Number of nodes in the first set |
np |
Number of nodes in the second set |
ties |
Number of ties; if this value is between 0 and 1, a random network where each tie is based on this probability will be produced |
weights |
A tie weight vector to be randomly sampled. If set to 1 (default), all tie weights will be 1, and hence a binary two-mode network will be created. |
seed |
the random seed. If you want it to be non-reproducable, use NULL otherwise, use a number |
Returns a random two-mode network
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Tore Opsahl. Triadic closure in two-mode networks: Redefining the global and local clustering coefficients. arXiv:1006.0887
## Run the programme rg_tm(ni=10,np=10,ties=20)## Run the programme rg_tm(ni=10,np=10,ties=20)
This function creates a classical random network with random edge weights.
rg_w(nodes=100,arcs=300,weights=1,directed=TRUE,seed=NULL)rg_w(nodes=100,arcs=300,weights=1,directed=TRUE,seed=NULL)
nodes |
number of nodes |
arcs |
number of arcs; if this value is between 0 and 1, a random network where each tie is based on this probability will be produced |
weights |
A tie weight vector to be randomly sampled. |
directed |
whether you want a directed or undirected network, values TRUE or FALSE |
seed |
the random seed. If you want it to be non-reproducable, use NULL otherwise, use a number |
Returns a one-mode network with random weights.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/tnet/
rg_w(nodes=10,arcs=30,directed=FALSE,seed=1)rg_w(nodes=10,arcs=30,directed=FALSE,seed=1)
This function creates a weighted edgelist from a list of edges where a duplicate means an increase in the weight.
shrink_to_weighted_network(net)shrink_to_weighted_network(net)
net |
can use both undirected and directed edgelist in the following format (sender.id receiver.id): |
Returns a weighted one-mode network, e.g.,
1 2 4
1 3 2
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2008/11/28/network-weighted-network/
## Load sample data sample <- rbind( c(1,2), c(1,2), c(1,2), c(1,2), c(1,3), c(1,3), c(2,1), c(2,1), c(2,1), c(2,1), c(2,3), c(2,3), c(2,3), c(2,3), c(2,4), c(2,5), c(2,5), c(3,1), c(3,1), c(3,2), c(3,2), c(3,2), c(3,2), c(4,2), c(5,2), c(5,2), c(5,6), c(6,5)) ## Run the programme shrink_to_weighted_network(sample)## Load sample data sample <- rbind( c(1,2), c(1,2), c(1,2), c(1,2), c(1,3), c(1,3), c(2,1), c(2,1), c(2,1), c(2,1), c(2,3), c(2,3), c(2,3), c(2,3), c(2,4), c(2,5), c(2,5), c(3,1), c(3,1), c(3,2), c(3,2), c(3,2), c(3,2), c(4,2), c(5,2), c(5,2), c(5,6), c(6,5)) ## Run the programme shrink_to_weighted_network(sample)
The symmetrise_w-function creates an undirected one-mode network from a directed one-mode network.
symmetrise_w(net, method="MAX")symmetrise_w(net, method="MAX")
net |
A one-mode network |
method |
the method used to decide the weight of the undirected edge. It can be: "MAX" sets the weight to the maximum of the weight(s) of the arc(s) "MIN" sets the weight to the minimumof the weight(s) of the arc(s) "AMEAN" sets the weight to the average (arithmetic mean) of the weight(s) of the arc(s) "SUM" sets the weight to the sum of the weight(s) of the arc(s) "PROD" sets the weight to the product of the weight(s) of the arc(s) "DIFF" sets the weight to the absolute difference between the weight(s) of the arc(s) |
Returns the undirected network
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/2008/11/28/network-weighted-network/
## Load sample data sample <- rbind( c(1,2,2), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(5,2,2), c(5,6,1)) ## Run the programme symmetrise_w(sample, method="MAX")## Load sample data sample <- rbind( c(1,2,2), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(5,2,2), c(5,6,1)) ## Run the programme symmetrise_w(sample, method="MAX")
The tnet_igraph function creates an igraph object from a tnet network.
tnet_igraph(net,type=NULL, directed=NULL)tnet_igraph(net,type=NULL, directed=NULL)
net |
A tnet network |
type |
type of tnet network, see as.tnet. |
directed |
if a one-mode networks, this can be set to avoid testing whether the network is directed. |
Returns the igraph object.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/
## Load sample data sample <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme tnet_igraph(sample, type="weighted one-mode tnet")## Load sample data sample <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme tnet_igraph(sample, type="weighted one-mode tnet")
The tnet_sna function creates a DL file which can easily be imported into UCINET.
tnet_ucinet(net, type=NULL, file=NULL)tnet_ucinet(net, type=NULL, file=NULL)
net |
A tnet network |
type |
type of tnet network, see as.tnet. |
file |
filename of output file. If this is set to NULL, a file is created in the working directory with the current time (e.g., tnet_ucinet_network-2011-04-10_150817.dl). |
Writes a UCINET dl file.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
http://toreopsahl.com/
## Load sample data sample <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme tnet_ucinet(sample, type="weighted one-mode tnet")## Load sample data sample <- rbind( c(1,2,4), c(1,3,2), c(2,1,4), c(2,3,4), c(2,4,1), c(2,5,2), c(3,1,2), c(3,2,4), c(4,2,1), c(5,2,2), c(5,6,1), c(6,5,1)) ## Run the programme tnet_ucinet(sample, type="weighted one-mode tnet")
The nodes in this network is the 500 busiest commercial airports in the United States. A tie exists between two airports if a flight was scheduled between them in 2002. The weights corresponds to the number of seats available on the scheduled flights. Even thought this type of networks is directed by nature as a flight is scheduled from one airport and to another, the networks are highly symmetric (Barrat et al., 2004). Therefore, the version of this network is undirected (i.e., the weight of the tie from one airport towards another is equal to the weight of the reciprocal tie). This network was obtained from the Complex Networks Collaboratory's website
See http://toreopsahl.com/datasets/
USairport.n500.netUSairport.n500.net
A data frame with three columns. The first two columns are the nodes' ids, and the third column is the weight of the tie.
Colizza, V., Pastor-Satorras, R., Vespignani, A., 2007. Reaction-diffusion processes and metapopulation models in heterogeneous networks. Nature Physics 3, 276-282.
See http://toreopsahl.com/datasets/
This function calculates the local weighted rich-club coefficient proposed in Opsahl, T., 2008. Local weighted rich-club measure.
http://toreopsahl.com/2008/12/26/local-weighted-rich-club-measure/
weighted_richclub_local_w(net, prominence)weighted_richclub_local_w(net, prominence)
net |
A weighted edgelist |
prominence |
A vector with 1 denoting prominent, and 0 non-prominent. This list must be as long as the highest node id number. |
Returns a table with the fraction of phi(observed) over phi(null) for each k or s in the dataset.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Opsahl et al., 2008. Prominence and control: The weighted rich-club effect. PRL 101
http://toreopsahl.com/2008/12/12/article-prominence-and-control-the-weighted-rich-club-effect/
## Load sample data sample <- cbind( i=c(1,1,2,2,2,2,3,3,4,5,5,6), j=c(2,3,1,3,4,5,1,2,2,2,6,5), w=c(4,2,4,4,1,2,2,4,1,2,1,1)) prominence <- c(1,1,1,0,0,0) ## Run the function weighted_richclub_local_w(sample, prominence)## Load sample data sample <- cbind( i=c(1,1,2,2,2,2,3,3,4,5,5,6), j=c(2,3,1,3,4,5,1,2,2,2,6,5), w=c(4,2,4,4,1,2,2,4,1,2,1,1)) prominence <- c(1,1,1,0,0,0) ## Run the function weighted_richclub_local_w(sample, prominence)
This function calculates the weighted rich-club coefficient proposed in Opsahl, T., Colizza, V., Panzarasa, P., Ramasco, J.J., 2008. Prominence and control: The weighted rich-club effect. PRL 101. It incorporates two extentions explained in this blog post http://toreopsahl.com/2009/05/29/weighted-rich-club-effect-a-more-appropriate-null-model-for-scientific-collaboration-networks/:
1) a new way of reshuffling (two-mode link reshuffling;
2) calculating significance levels if there are more than 100 random networks (see my PhD thesis; http://toreopsahl.com/publications/thesis/)
weighted_richclub_tm(net, NR=1000, seed=NULL, projection.method="Newman", nbins=30)weighted_richclub_tm(net, NR=1000, seed=NULL, projection.method="Newman", nbins=30)
net |
A binary two-mode edgelist |
NR |
number of random networks used. |
seed |
the random generators seed, used to produce random yet reproducable results. |
projection.method |
the method used to project the two-mode network to a weighted one-mode network: either "sum" or "Newman" |
nbins |
the number of bins in the output |
Returns a table with the fraction of phi(observed) over phi(null). Nbins controls the number of rows.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Opsahl et al., 2008. Prominence and control: The weighted rich-club effect. PRL 101
http://toreopsahl.com/2008/12/12/article-prominence-and-control-the-weighted-rich-club-effect/
http://toreopsahl.com/2009/05/29/weighted-rich-club-effect-a-more-appropriate-null-model-for-scientific-collaboration-networks/
## Load data data(tnet) ## Run the function on a subset weighted_richclub_tm(Newman.Condmat.95.99.net.2mode[1:100,], NR=10)## Load data data(tnet) ## Run the function on a subset weighted_richclub_tm(Newman.Condmat.95.99.net.2mode[1:100,], NR=10)
This function calculates the weighted rich-club coefficient proposed in Opsahl, T., Colizza, V., Panzarasa, P., Ramasco, J.J., 2008. Prominence and control: The weighted rich-club effect. PRL 101.
http://toreopsahl.com/2008/12/12/article-prominence-and-control-the-weighted-rich-club-effect/
weighted_richclub_w(net, rich="k", reshuffle="weights", NR=1000, nbins=30, seed=NULL, directed=NULL)weighted_richclub_w(net, rich="k", reshuffle="weights", NR=1000, nbins=30, seed=NULL, directed=NULL)
net |
A weighted edgelist |
rich |
specifies the richness parameter, either "k" or "s". |
reshuffle |
specifies the reshuffling procedure used, either "weights" or "links". |
NR |
number of random networks used. |
nbins |
the number of bins in the output |
seed |
the random generators seed, used to produce random yet reproducable results. |
directed |
logical parameter: whether the network is directed or undirected. |
Returns a table with the fraction of phi(observed) over phi(null) for each k or s in the dataset.
version 1.0.0
Tore Opsahl; http://toreopsahl.com
Opsahl et al., 2008. Prominence and control: The weighted rich-club effect. PRL 101
http://toreopsahl.com/2008/12/12/article-prominence-and-control-the-weighted-rich-club-effect/
## Load sample data sample <- cbind( i=c(1,1,2,2,2,2,3,3,4,5), j=c(2,3,1,3,4,5,1,2,2,2), w=c(4,2,4,4,1,2,2,4,1,2)) ## Run the function weighted_richclub_w(sample, rich="k", reshuffle="weights", NR = 100)## Load sample data sample <- cbind( i=c(1,1,2,2,2,2,3,3,4,5), j=c(2,3,1,3,4,5,1,2,2,2), w=c(4,2,4,4,1,2,2,4,1,2)) ## Run the function weighted_richclub_w(sample, rich="k", reshuffle="weights", NR = 100)