Package 'SemNeT' reference manual

Title:	Methods and Measures for Semantic Network Analysis
Description:	Implements several functions for the analysis of semantic networks including different network estimation algorithms, partial node bootstrapping (Kenett, Anaki, & Faust, 2014 <doi:10.3389/fnhum.2014.00407>), random walk simulation (Kenett & Austerweil, 2016 <http://alab.psych.wisc.edu/papers/files/Kenett16CreativityRW.pdf>), and a function to compute global network measures. Significance tests and plotting features are also implemented.
Authors:	Alexander P. Christensen [aut, cre] , Yoed N. Kenett [aut, ctb]
Maintainer:	Alexander P. Christensen <[email protected]>
License:	GPL (>= 3.0)
Version:	1.4.4
Built:	2024-11-05 03:27:23 UTC
Source:	https://github.com/alexchristensen/semnet

SemNeT–package

Description

Implements several functions for the analysis of semantic networks including partial node bootstraping (Kenett, Anaki, & Faust, 2014), random walk simulation (Kenett & Austerweil, 2016), and a function to compute global network measures. Significance tests and plotting features are also implemented.

Author(s)

Alexander P. Christensen <[email protected]> & Yoed N. Kenett <[email protected]>

References

Christensen, A. P., Kenett, Y. N., Cotter, K. N., Beaty, R. E., & Silvia, P. J. (2018). Remotely close associations: Openness to experience and semantic memory structure. European Journal of Personality, 32, 480-492.

Kenett, Y. N., Anaki, D., & Faust, M. (2014). Investigating the structure of semantic networks in low and high creative persons. Frontiers in Human Neuroscience, 8, 407.

Kenett, Y. N., & Austerweil, J. L. (2016). Examining search processes in low and high creative individuals with random walks. In Paper presented at the proceedings of the 38th annual meeting of the cognitive science society. Austin, TX.

Frequency of Animal Responses

Description

Frequency of animal responses from Christensen & Kenett (2019). These frequencies are used to generate data in the sim.fluency function.

Usage

data(animals.freq)
data(animals.freq)

Format

animals.freq (vector, length = 367)

Examples

data("animals.freq")
data("animals.freq")

Average Shortest Path Length

Description

Computes the global average shortest path length of the network

Usage

ASPL(A, weighted = FALSE)
ASPL(A, weighted = FALSE)

Arguments

`A`	An adjacency matrix of network data
`weighted`	Is the network weighted? Defaults to `FALSE`. Set to `TRUE` for weighted measures

Value

Returns the ASPL of the network

Author(s)

Alexander Christensen <[email protected]>

References

Rubinov, M., & Sporns, O. (2010). Complex network measures of brain connectivity: Uses and interpretations. NeuroImage, 52, 1059-1069.

Examples

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(100), method = "cor"))

# Unweighted
aspl <- ASPL(A)

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(100), method = "cor"))

# Unweighted
aspl <- ASPL(A)

Bootstrapped Semantic Network Analysis

Description

Bootstrap techniques to generate semantic networks and compute global network characteristics

Usage

bootSemNeT(
  ...,
  method = c("CN", "NRW", "PF", "TMFG"),
  methodArgs = list(),
  type = c("case", "node"),
  prop = 0.5,
  sim,
  weighted = FALSE,
  iter = 1000,
  cores
)
bootSemNeT(
  ...,
  method = c("CN", "NRW", "PF", "TMFG"),
  methodArgs = list(),
  type = c("case", "node"),
  prop = 0.5,
  sim,
  weighted = FALSE,
  iter = 1000,
  cores
)

Arguments

`...`	Matrices or data frames. Cleaned response matrices (e.g., `responses$clean` from `textcleaner`) or binary response matrices (e.g., `binary` output from `textcleaner`)
`method`	Character. Network estimation method to use. Current options include: `TMFG` Triangulated Maximally Filtered Graph `CN` Community Network `NRW` Naive Random Walk `PF` Pathfinder
`methodArgs`	List. A list of additional arguments for the network estimation function. See links in argument `method` for additional arguments (see also Examples)
`type`	Character. Type of bootstrap to perform `node` Generates partial networks based on dropping a certain proportion of nodes (see argument `prop`) `case` Samples with replacement the same number of participants as in the original dataset
`prop`	Numeric. Only for `type = "node"`. Proportion of nodes to remain in the network. Defaults to `.50`
`sim`	Character. Similarity measure to use. Defaults to `"cosine"`. See `similarity` for other options
`weighted`	Boolean. Should weighted ASPL and CC be used? Defaults to `FALSE`. Set to `TRUE` for weighted ASPL and CC
`iter`	Numeric. Number of iterations in bootstrap. Defaults to `1000`
`cores`	Numeric. Number of computer processing cores to use for bootstrapping samples. Defaults to n / 2 total number of cores. Set to any number between 1 and maximum amount of cores on your computer (see `parellel::detectCores()`)

Value

Returns a list containing:

`dataMeas`	A matrix for the network input in the `data` argument, where columns are the semantic network measures from `semnetmeas` and rows are their values from each bootstrapped sample (results in a matrix with the dimensions `iter` by 3)
`dataSumm`	Summary statistics across the bootrapped samples for the network input in the `data` argument
`prop`	Outputs the proportion used from the `prop` argument
`iter`	Outputs the number of bootstrapped samples used from the `iter` argument

If a paired network is input, then also returns:

`pairedMeas`	A matrix for the network input in the `paired` argument, where columns are the semantic network measures from `semnetmeas` and rows are their values from each bootstrapped sample (results in a matrix with the dimensions `iter` by 3)
`pairedSumm`	Summary statistics across the bootrapped samples for the network input in the `paired` argument

Author(s)

Alexander Christensen <[email protected]>

Examples

# Simulate Dataset
one <- sim.fluency(20)

# Run bootstrap node-drop (partial) networks
one.result <- bootSemNeT(one, prop = .50, iter = 100,
sim = "cosine", cores = 2, method = "TMFG", type = "node")

# Run bootstrap case-drop networks
## Includes additional equating argument: minCase
one.result <- bootSemNeT(one, iter = 100, sim = "cosine",
cores = 2, method = "TMFG", type = "case", methodArgs = list(minCase = 2))


# Bootstrap case-wise networks
## Get openness data
low <- open.clean[which(open.group == "Low"),]
high <- open.clean[which(open.group == "High"),]

## Run
### Inlcudes additional NRW argument: threshold
open <- bootSemNeT(low, high, iter = 100, cores = 2, method = "NRW", type = "case",
methodArgs = list(type = "num", threshold = 3))


# Simulate Dataset
one <- sim.fluency(20)

# Run bootstrap node-drop (partial) networks
one.result <- bootSemNeT(one, prop = .50, iter = 100,
sim = "cosine", cores = 2, method = "TMFG", type = "node")

# Run bootstrap case-drop networks
## Includes additional equating argument: minCase
one.result <- bootSemNeT(one, iter = 100, sim = "cosine",
cores = 2, method = "TMFG", type = "case", methodArgs = list(minCase = 2))


# Bootstrap case-wise networks
## Get openness data
low <- open.clean[which(open.group == "Low"),]
high <- open.clean[which(open.group == "High"),]

## Run
### Inlcudes additional NRW argument: threshold
open <- bootSemNeT(low, high, iter = 100, cores = 2, method = "NRW", type = "case",
methodArgs = list(type = "num", threshold = 3))

Clustering Coefficient

Description

Computes global clustering coefficient CC

Usage

CC(A, weighted = FALSE)
CC(A, weighted = FALSE)

Arguments

`A`	An adjacency matrix of network data
`weighted`	Is the network weighted? Defaults to `FALSE`. Set to `TRUE` for weighted measures of CC and CCi

Value

Returns the network's CC

Author(s)

Alexander Christensen <[email protected]>

References

Rubinov, M., & Sporns, O. (2010). Complex network measures of brain connectivity: Uses and interpretations. NeuroImage, 52, 1059-1069.

Examples

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(100), method = "cor"))

# Unweighted
cc <- CC(A)

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(100), method = "cor"))

# Unweighted
cc <- CC(A)

Community Network Estimation

Description

Estimates a semantic network using the Community Network method described in Goni et al. (2011)

Usage

CN(data, window = 2, alpha = 0.05, enrich = FALSE)
CN(data, window = 2, alpha = 0.05, enrich = FALSE)

Arguments

`data`	Matrix or data frame. A preprocessed verbal fluency matrix where rows are participants and columns are verbal fluency responses
`window`	Numeric. Size of window to look for co-occurences in. Defaults to `2`
`alpha`	Numeric. Significance value. Defaults to `.05`
`enrich`	Boolean. Should the network be enriched by connecting all nodes in their respective modules? Defaults to `FALSE`

Value

Returns a undirected semantic network

Author(s)

Alexander Christensen <[email protected]>

References

Goni, J., Arrondo, G., Sepulcre, J., Martincorena, I., de Mendizabal, N. V., Corominas-Murtra, B., ... & Villoslada, P. (2011). The semantic organization of the animal category: Evidence from semantic verbal fluency and network theory. Cognitive Processing, 12, 183-196.

Examples

# Get data
data <- open.clean

# Organize group data
## Get group data
group <- open.group

## Low and high openness to experience groups
low <- data[which(group == "Low"),]
high <- data[which(group == "High"),]

## Not run: 
# Compute networks
low.net <- CN(low)
high.net <- CN(high)

## End(Not run)

# Get data
data <- open.clean

# Organize group data
## Get group data
group <- open.group

## Low and high openness to experience groups
low <- data[which(group == "Low"),]
high <- data[which(group == "High"),]

## Not run: 
# Compute networks
low.net <- CN(low)
high.net <- CN(high)

## End(Not run)

Plots Networks for Comparison

Description

Uses qgraph to plot networks. Accepts any number of networks and will organize the plots in the number of side-by-side columns using the heuristic of taking the square root of the number of input and rounding down to the nearest integer (i.e., floor(sqrt(length(input)))).

Examples

3 networks: 1 x 3
6 networks: 2 x 3
9 networks: 3 x 3

Usage

compare_nets(
  ...,
  title,
  config,
  placement = c("match", "default"),
  weighted = FALSE,
  qgraph.args = list()
)
compare_nets(
  ...,
  title,
  config,
  placement = c("match", "default"),
  weighted = FALSE,
  qgraph.args = list()
)

Arguments

`...`	Matrices or data frames of network adjacency matrices
`title`	List. Characters denoting titles of plots
`config`	Character. Defaults to `"spring"` See `qgraph` for more options
`placement`	Character. How should nodes be placed when comparing groups? Defaults to `"default"` `"match"` places nodes in the same position for all networks `"default"` places nodes in the default `config` positions
`weighted`	Boolean. Should networks be plotted with weights? Defaults to `FALSE`. Set to `TRUE` to plot networks with weights corresponding to association strength. Often, unweighted networks are more aesthetically representational of the networks
`qgraph.args`	List. An argument list to be passed onto `qgraph`. See `qgraph` for possible arguments

Value

Plots networks using qgraph

Author(s)

Alexander Christensen <[email protected]>

References

Epskamp, S., Cramer, A. O. J., Waldorp, L. J., Schmittmann, V. D., & Borsboom, D. (2012). qgraph: Network visualizations of relationships in psychometric data. Journal of Statistical Software, 48, 1-18.

Jones, P. J. (2019). networktools: Tools for Identifying Important Nodes in Networks. R package version 1.2.1.

Jones, P. J., Mair, P., & McNally, R. (2018). Visualizing psychological networks: A tutorial in R. Frontiers in Psychology, 9, 1742.

Examples

# Simulate Datasets
one <- sim.fluency(10)
two <- sim.fluency(10)

# Compute similarity matrix
cos1 <- similarity(one, method = "cosine")
cos2 <- similarity(two, method = "cosine")

# Compute networks
net1 <- TMFG(cos1)
net2 <- TMFG(cos2)

# Compare networks
compare_nets(net1, net2, title = list("One", "Two"), config = "spring")

# Change edge colors
compare_nets(net1, net2, title = list("One", "Two"),
config = "spring", qgraph.args = list(edge.color = "blue"))

# Simulate Datasets
one <- sim.fluency(10)
two <- sim.fluency(10)

# Compute similarity matrix
cos1 <- similarity(one, method = "cosine")
cos2 <- similarity(two, method = "cosine")

# Compute networks
net1 <- TMFG(cos1)
net2 <- TMFG(cos2)

# Compare networks
compare_nets(net1, net2, title = list("One", "Two"), config = "spring")

# Change edge colors
compare_nets(net1, net2, title = list("One", "Two"),
config = "spring", qgraph.args = list(edge.color = "blue"))

Convert Adjacency Matrix to Cytoscape Format

Description

Converts an adjacency matrix to Cytoscape's sparse matrix format

Usage

convert2cytoscape(A)
convert2cytoscape(A)

Arguments

`A`	Matrix or data frame. A cleaned, finalized response matrix ready to be visualized

Value

A sparse matrix formatted for Cytoscape

Author(s)

Alexander Christensen <[email protected]>

References

Shannon, P., Markiel, A., Ozier, O., Baliga, N. S., Wang, J. T., Ramage, D., ... & Ideker, T. (2003). Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Research, 13, 2498-2504.

Examples

# Simulate Datasets
one <- sim.fluency(10)
two <- sim.fluency(10)

# Compute similarity matrix
cos1 <- similarity(one, method = "cosine")
cos2 <- similarity(two, method = "cosine")

# Compute networks
net1 <- TMFG(cos1)
net2 <- TMFG(cos2)

# Convert to Cytoscape format
cyto1 <- convert2cytoscape(net1)
cyto2 <- convert2cytoscape(net2)

# Write to .csv
write.csv(cyto1, file.path(tempdir(), "cyto1.csv"), row.names = FALSE)
write.csv(cyto2, file.path(tempdir(), "cyto2.csv"), row.names = FALSE)

# Simulate Datasets
one <- sim.fluency(10)
two <- sim.fluency(10)

# Compute similarity matrix
cos1 <- similarity(one, method = "cosine")
cos2 <- similarity(two, method = "cosine")

# Compute networks
net1 <- TMFG(cos1)
net2 <- TMFG(cos2)

# Convert to Cytoscape format
cyto1 <- convert2cytoscape(net1)
cyto2 <- convert2cytoscape(net2)

# Write to .csv
write.csv(cyto1, file.path(tempdir(), "cyto1.csv"), row.names = FALSE)
write.csv(cyto2, file.path(tempdir(), "cyto2.csv"), row.names = FALSE)

Convert Network(s) to igraph's Format

Description

Converts single or multiple networks into igraph's format for network analysis

Usage

convert2igraph(A, neural = FALSE)
convert2igraph(A, neural = FALSE)

Arguments

`A`	Adjacency matrix (network matrix) or brain connectivity array (from `convertConnBrainMat`)
`neural`	Defunct.

Value

Returns a network matrix in igraph's format

Author(s)

Alexander Christensen <[email protected]>

Examples

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(50), method = "cor"))

igraphNetwork <- convert2igraph(A)

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(50), method = "cor"))

igraphNetwork <- convert2igraph(A)

Equate Groups

Description

A function to "equate" multiple response matrices to one another. N number of groups are matched based on their responses so that every group has the same responses in their data

Usage

equate(...)
equate(...)

Arguments

...

Matrices, data frames or a list of matrices and data frames. Binary response matrices to be equated

Value

This function returns a list containing the equated binary response matrices in the order they were input. The response matrices are labeled as the object name they were entered with

Author(s)

Alexander Christensen <[email protected]>

Examples

# Obtain binary data
bin <- open.binary

# Finalize mat1
mat1 <- finalize(bin[c(1:5),])

# Finalize mat2
mat2 <- finalize(bin[c(6:10),])

# Equate mat1 and mat1
eq <- equate(mat1, mat2)

# Obtain respective equated response matrices
eq.mat1 <- eq$mat1 # list objects are named with the names
eq.mat2 <- eq$mat2 # they were entered with

# Obtain binary data
bin <- open.binary

# Finalize mat1
mat1 <- finalize(bin[c(1:5),])

# Finalize mat2
mat2 <- finalize(bin[c(6:10),])

# Equate mat1 and mat1
eq <- equate(mat1, mat2)

# Obtain respective equated response matrices
eq.mat1 <- eq$mat1 # list objects are named with the names
eq.mat2 <- eq$mat2 # they were entered with

Finalize Response Matrix

Description

Finalizes the response matrix by keeping responses that are given by a certain number of people

Usage

finalize(rmat, minCase = 2)
finalize(rmat, minCase = 2)

Arguments

`rmat`	Binary matrix. A textcleaner filtered response matrix
`minCase`	Numeric. Minimum number of cases to produce a response

Value

A binary response matrix with responses given by at least minCase people

Author(s)

Alexander Christensen <[email protected]>

Examples

# Obtain binary data
bin <- open.binary

# Finalize mat1
mat1 <- finalize(bin)

# Obtain binary data
bin <- open.binary

# Finalize mat1
mat1 <- finalize(bin)

High Openness to Experience Network

Description

High openness to experience network from Christensen & Kenett (2019)

Usage

data(net.high)
data(net.high)

Format

net.high (matrix, 160 x 160)

References

Christensen, A. P., & Kenett, Y. N. (2019) Semantic network analysis (SemNA): A tutorial on preprocessing, estimating, and analyzing semantic networks. PsyArXiv.

Examples

data("net.high")
data("net.high")

Low Openness to Experience Network

Description

Low openness to experience network from Christensen & Kenett (2019)

Usage

data(net.low)
data(net.low)

Format

net.low (matrix, 160 x 160)

References

Christensen, A. P., & Kenett, Y. N. (2019) Semantic network analysis (SemNA): A tutorial on preprocessing, estimating, and analyzing semantic networks. PsyArXiv.

Examples

data("net.low")
data("net.low")

Naive Random Walk Network Estimation

Description

Estimates a semantic network using the Naive Random Walk method described in Lerner, Ogrocki, and Thomas (2009)

Usage

NRW(data, type = c("num", "prop"), threshold = 0)
NRW(data, type = c("num", "prop"), threshold = 0)

Arguments

data

Matrix or data frame. A preprocessed verbal fluency matrix where rows are participants and columns are verbal fluency responses

type

Character. Type of threshold to apply.

"num" Minimum number of co-occurrences
"prop" Minimum proportion of co-occurrences

Defaults to "num"

threshold

Numeric. Value of the minimum number or proportion of co-occurrences. Defaults to 0 for both "num" and "prop"

Value

Returns a undirected semantic network

Author(s)

Alexander Christensen <[email protected]>

References

Lerner, A. J., Ogrocki, P. K., & Thomas, P. J. (2009). Network graph analysis of category fluency testing. Cognitive and Behavioral Neurology, 22, 45-52.

Examples

# Get data
data <- open.clean

# Organize group data
## Get group data
group <- open.group

## Low and high openness to experience groups
low <- data[which(group == "Low"),]
high <- data[which(group == "High"),]

# Compute networks
low.net <- NRW(low)
high.net <- NRW(high)

# Get data
data <- open.clean

# Organize group data
## Get group data
group <- open.group

## Low and high openness to experience groups
low <- data[which(group == "Low"),]
high <- data[which(group == "High"),]

# Compute networks
low.net <- NRW(low)
high.net <- NRW(high)

Simulated Result for Dataset One

Description

A result of bootSemNeT from a simulated dataset

Usage

data(one.result)
data(one.result)

Format

one.result (list, length = 4)

Examples

data("one.result")
data("one.result")

Binary response Matrices (Openness and Verbal Fluency)

Description

Binary response matrices for the Animals verbal fluency data (n = 516) from Christensen et al. (2018).

Usage

data(open.binary)
data(open.binary)

Format

open.binary (matrix, 516 x 367)

References

Examples

data("open.binary")
data("open.binary")

Cleaned response Matrices (Openness and Verbal Fluency)

Description

Cleaned response matrices for the Animals verbal fluency data (n = 516) from Christensen et al. (2018).

Usage

data(open.clean)
data(open.clean)

Format

open.clean (matrix, 516 x 35)

References

Examples

data("open.clean")
data("open.clean")

Groups for Openness and Verbal Fluency

Description

Groups for the Animals verbal fluency data (n = 516) from Christensen et al. (2018; see also open.clean).

Usage

data(open.group)
data(open.group)

Format

open.group (vector, length = 516)

References

Examples

data("open.group")
data("open.group")

Pathfinder Network

Description

Estimates a pathfinder network using the MST-Pathfinder Network method from Quirin et al. (2008; see also Schvaneveldt, 1990)

Usage

PF(data)
PF(data)

Arguments

data

Matrix or data frame. A binary response matrix

Value

An adjacency matrix

Author(s)

Alexander Christensen <[email protected]>

References

Quirin, A., Cordon, O., Guerrero-Bote, V. P., Vargas-Quesada, B., & Moya-Aneon, F. (2008) A quick MST-based algorithm to obtain Pathfinder networks (Inf, n-1). Journal of the American Society for Information Science and Technology, 59, 1912-1924.

Schvaneveldt, R. W. (1990). Pathfinder associative networks: Studies in knowledge organization. Norwood, NJ: Ablex Publishing.

Examples

# Obtain data
data <- open.binary

# Estimate network
pf.net <- PF(data)

# Obtain data
data <- open.binary

# Estimate network
pf.net <- PF(data)

Animate Networks for Spreading Activation from Shiny

Description

Uses qgraph and ani.record to animate networks. Accepts only one network animation at a time

Usage

## S3 method for class 'animateShiny'
plot(x, ...)
## S3 method for class 'animateShiny'
plot(x, ...)

Arguments

`x`	Shiny result `resultShiny$spreadingActivationPlot`
`...`	Additional arguments for `ani.record`

Value

Plots animated networks using qgraph and ani.record

Author(s)

Alexander Christensen <[email protected]>

References

Siew, C. S. Q. (2019). spreadr: An R package to simulate spreading activation in a network. Behavior Research Methods, 51, 910-929. https://doi.org/10.3758/s13428-018-1186-5

Examples


if(interactive())
{SemNeTShiny()}

## Not run: 
  plot(resultShiny$spreadingActivationPlot[[1]])

## End(Not run)

if(interactive())
{SemNeTShiny()}

## Not run: 
  plot(resultShiny$spreadingActivationPlot[[1]])

## End(Not run)

Plot for bootSemNeT

Description

Plots output from bootSemNeT

Usage

## S3 method for class 'bootSemNeT'
plot(..., groups = NULL, measures = c("ASPL", "CC", "Q"))
## S3 method for class 'bootSemNeT'
plot(..., groups = NULL, measures = c("ASPL", "CC", "Q"))

Arguments

`...`	Object(s) from `bootSemNeT`
`groups`	Character. Labels for groups in the order they were entered in `bootSemNeT`
`measures`	Character. Measures to be plotted

Value

Returns plots for the specified measures

Author(s)

Alexander Christensen <[email protected]>

Examples

# Simulate Dataset
one <- sim.fluency(20)

# Run partial bootstrap networks
one.result <- bootSemNeT(one, prop = .50, iter = 1000,
sim = "cosine", cores = 2, type = "node", method = "TMFG")

# Plot
plot(one.result, groups = c("One"))

# Simulate Dataset
one <- sim.fluency(20)

# Run partial bootstrap networks
one.result <- bootSemNeT(one, prop = .50, iter = 1000,
sim = "cosine", cores = 2, type = "node", method = "TMFG")

# Plot
plot(one.result, groups = c("One"))

Plots Networks for Comparison from Shiny

Description

Examples

3 networks: 1 x 3
6 networks: 2 x 3
9 networks: 3 x 3

Usage

## S3 method for class 'compareShiny'
plot(x, ...)
## S3 method for class 'compareShiny'
plot(x, ...)

Arguments

`x`	Shiny result `resultShiny$comparePlot`
`...`	Additional arguments

Value

Plots networks using qgraph

Author(s)

Alexander Christensen <[email protected]>

References

Jones, P. J. (2019). networktools: Tools for Identifying Important Nodes in Networks. R package version 1.2.1.

Jones, P. J., Mair, P., & McNally, R. (2018). Visualizing psychological networks: A tutorial in R. Frontiers in Psychology, 9, 1742.

Examples

# Simulate Datasets
one <- sim.fluency(10)
two <- sim.fluency(10)

# Compute similarity matrix
cos1 <- similarity(one, method = "cosine")
cos2 <- similarity(two, method = "cosine")

# Compute networks
net1 <- TMFG(cos1)
net2 <- TMFG(cos2)

# Compare networks
compare_nets(net1, net2, title = list("One", "Two"), config = "spring")

# Change edge colors
compare_nets(net1, net2, title = list("One", "Two"),
config = "spring", qgraph.args = list(edge.color = "blue"))

# Simulate Datasets
one <- sim.fluency(10)
two <- sim.fluency(10)

# Compute similarity matrix
cos1 <- similarity(one, method = "cosine")
cos2 <- similarity(two, method = "cosine")

# Compute networks
net1 <- TMFG(cos1)
net2 <- TMFG(cos2)

# Compare networks
compare_nets(net1, net2, title = list("One", "Two"), config = "spring")

# Change edge colors
compare_nets(net1, net2, title = list("One", "Two"),
config = "spring", qgraph.args = list(edge.color = "blue"))

Modularity

Description

Computes a global modularity measure (Q) using the Louvain community detection algorithm

Usage

Q(A)
Q(A)

Arguments

`A`	An adjacency matrix of network data

Value

Returns Q or a measure of how well the communities in the network are compartmentalized

Author(s)

Alexander Christensen <[email protected]>

References

Blondel, V. D., Guillaume, J. L., Lambiotte, R., & Lefebvre, E. (2008). Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment, 2008, P10008.

Rubinov, M., & Sporns, O. (2010). Complex network measures of brain connectivity: Uses and interpretations. NeuroImage, 52, 1059-1069.

Examples

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(100), method = "cor"))

modularity <- Q(A)

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(100), method = "cor"))

modularity <- Q(A)

Test Against Random Networks

Description

Performs significance tests for global measures of semantic networks against the global measures of equivalent size (and density) random networks

Usage

randnet.test(..., iter, cores)
randnet.test(..., iter, cores)

Arguments

`...`	Matrices or data frames. Semantic networks to be compared against random networks
`iter`	Numeric. Number of iterations in bootstrap. Defaults to `1000`
`cores`	Number of computer processing cores to use for bootstrapping samples. Defaults to n - 1 total number of cores. Set to any number between 1 and maximum amount of cores on your computer

Value

Returns a matrix containing p-values for the network measures of the input networks against the distribution of equivalent random networks. The last two columns contain the mean ("M.rand") and standard deviation ("SD.rand") of the network measures for the random network distribution

Author(s)

Alexander Christensen <[email protected]>

References

Viger, F., & Latapy, M. (2016). Efficient and simple generation of random simple connected graphs with prescribed degree sequence. Journal of Complex Networks, 4, 15-37.

Examples

# Get openness data
one <- open.clean[which(open.group == "Low"),]
two <- open.clean[which(open.group == "High"),]

# Compute networks
net.one <- CN(one)
net.two <- CN(two)

# Perform random networks test
randnet.test(net.one, net.two, iter = 100, cores = 2)


# Get openness data
one <- open.clean[which(open.group == "Low"),]
two <- open.clean[which(open.group == "High"),]

# Compute networks
net.one <- CN(one)
net.two <- CN(two)

# Perform random networks test
randnet.test(net.one, net.two, iter = 100, cores = 2)

Random Walk Simulation

Description

Simulates random walks over two networks to examine the characteristics of spontaneous spreading activation (see Kenett & Austerweil, 2016)

Usage

randwalk(A, B, reps = 20, steps = 10, iter = 10000, cores)
randwalk(A, B, reps = 20, steps = 10, iter = 10000, cores)

Arguments

`A`	Matrix or data frame. Adjacency matrix of a semantic network
`B`	Matrix or data frame. A comparison adjacency matrix of a semantic network
`reps`	Numeric. Number of repetitions of increments in 10 steps. Defaults to `20`
`steps`	Numeric. Number of random steps to begin with. Defaults to `10`
`iter`	Numeric. Number of iterations for each random walk. Defaults to `10000`
`cores`	Numeric. Number of computer processing cores to use for bootstrapping samples. Defaults to n - 1 total number of cores. Set to any number between 1 and maximum amount of cores on your computer

Value

A result matrix containing the means and standard deviations for several measures as well as p-values for a Mann-Whitney U test

Author(s)

Alexander Christensen <[email protected]> and Yoed Kenett <[email protected]>

References

Examples

# Simulate Datasets
one <- sim.fluency(10)
two <- sim.fluency(10)

# Compute similarity matrix
cos1 <- similarity(one, method = "cosine")
cos2 <- similarity(two, method = "cosine")

# Compute networks
net1 <- TMFG(cos1)
net2 <- TMFG(cos2)

# Run random walk analysis
rw.results <- randwalk(net1, net2, iter = 100, cores = 2)



# Simulate Datasets
one <- sim.fluency(10)
two <- sim.fluency(10)

# Compute similarity matrix
cos1 <- similarity(one, method = "cosine")
cos2 <- similarity(two, method = "cosine")

# Compute networks
net1 <- TMFG(cos1)
net2 <- TMFG(cos2)

# Run random walk analysis
rw.results <- randwalk(net1, net2, iter = 100, cores = 2)

Response Analysis

Description

Computes the difference in the total and unique number of responses between two groups (follows Christensen et al., 2018)

Usage

response.analysis(...)
response.analysis(...)

Arguments

...

Matrix or data frame. Responses matrices for two different groups

Value

A list containing objects:

`total`	A vector with the total responses given by each participant. A t-test is used to compare, on average, whether one group provides more response than the other
`unique`	A vector with the number of unique responses provided by both groups (`Total Across Groups`), the number of unique responses provided by each group (`Total`), and the number of unique resposnes provided by each group that were not provided by the other group (`Unique`). A McNemar's test is used to compare whether the number of unique responses are different between groups

Author(s)

Alexander Christensen <[email protected]>

References

Examples

# Obtain data
low <- open.clean[which(open.group == "Low"),]
high <- open.clean[which(open.group == "High"),]

# Perform analysis
response.analysis(low, high)

# Obtain data
low <- open.clean[which(open.group == "Low"),]
high <- open.clean[which(open.group == "High"),]

# Perform analysis
response.analysis(low, high)

Semantic Network Measures

Description

Computes the average shortest path length (ASPL), clustering coefficient(CC), and modularity (Q) of the network

Usage

semnetmeas(A, meas = c("ASPL", "CC", "Q"), weighted = FALSE)
semnetmeas(A, meas = c("ASPL", "CC", "Q"), weighted = FALSE)

Arguments

`A`	Matrix or data frame. An adjacency matrix of a network
`meas`	Character. Global network measures to compute. By default, computes ASPL, CC, and Q. Individual measures can be selected
`weighted`	Boolean. Should weighted measures be computed? Defaults to `FALSE`. Set to `TRUE` for weighted measures

Value

Returns a values for ASPL, CC, and Q

Author(s)

Alexander Christensen <[email protected]>

Examples

# Simulate Datasets
one <- sim.fluency(10)

# Compute similarity matrix
cos <- similarity(one, method = "cosine")

# Compute networks
net <- TMFG(cos)

# Compute global network measures
globmeas <- semnetmeas(net)

# Simulate Datasets
one <- sim.fluency(10)

# Compute similarity matrix
cos <- similarity(one, method = "cosine")

# Compute networks
net <- TMFG(cos)

# Compute global network measures
globmeas <- semnetmeas(net)

Shiny App for `SemNeT`

Description

An interactive Shiny application for running SemNeT analysis.

Usage

SemNeTShiny()
SemNeTShiny()

Value

A list called resultShiny containing:

`data`	The data imported into `SemNeTShiny`
`group`	The grouping variable imported into `SemNeTShiny`
`network`	The networks generated during `SemNeTShiny` session. The networks are labelled using the provided grouping variable
`measures`	Network measures ASPL (Average Shortest Path Lengths), CC (Clustering Coefficient), and Q (Modularity) for the networks generated during `SemNeTShiny`

(see semnetmeas)

comparePlot

A visualization of the networks generated during SemNeTShiny

(see compare_nets)

`randomTest`	Statistical results from the Random Network Test in `SemNeTShiny` (see `randnet.test`)
`bootstrap`	Results from the Bootstrap Network Analysis in `SemNeTShiny` (see `bootSemNeT`)
`bootstrapTest`	Statistical results from the Bootstrap Network Analysis (see `test.bootSemNeT`)
`bootstrapPlot`	Plots of the statistical results from the Bootstrap Network Analysis (see `plot.bootSemNeT`)
`randomWalk`	Results from the Random Walk Analysis (see `randwalk`)
`bootstrapPlot`	Results from the Spreading Activation Analysis (see `spreadr`)

Examples


if(interactive())
{SemNeTShiny()}


if(interactive())
{SemNeTShiny()}

Simulates a verbal fluency binary response matrix

Description

Simulates verbal fluency data based on the number of nodes in the desired network. The summed total of each response is simulated from a poisson distribution (see rpois), using frequencies from the animals.freq data. Using these sums, participants responses are simulated with a probability of giving a response as the total of the summed response over the number of participants.

Usage

sim.fluency(nodes, cases, random = FALSE)
sim.fluency(nodes, cases, random = FALSE)

Arguments

`nodes`	Numeric. Number of nodes to simulate in data. Defaults to `100`
`cases`	Numeric. Number of participants to simulate in data. Defaults to `500`
`random`	Boolean. Should the frequencies be randomly sampled from? Defaults to `FALSE`.

Value

A binary matrix with p (participants) by n (nodes)

Author(s)

Alexander Christensen <[email protected]>

Examples

# Simulate data for 50 nodes and 200 participants
sim.fluency(nodes = 50, cases = 200)

# Simulate data for 50 nodes and 200 participants
sim.fluency(nodes = 50, cases = 200)

Measures of Similarity

Description

Computes several measures of similarity (see Choi, Cha, & Tappert, 2010 for additional measures)

Usage

similarity(
  data,
  method = c("angular", "cor", "cosine", "euclid", "faith", "jaccard", "phi", "rr")
)
similarity(
  data,
  method = c("angular", "cor", "cosine", "euclid", "faith", "jaccard", "phi", "rr")
)

Arguments

data

Matrix or data frame. A binarized dataset of verbal fluency or linguistic data

method

Character. Type of similarity measure to compute.

Below are the definitions for each bin:

	1	0
1	a	b	a+b	(R1)
0	c	d	c+d	(R2)
	a+c	b+d	a+b+c+d	(N)
	(C1)	(C2)	(N)

Options include:

"angular" = $1 - (2 * acos(cosine similarity) / \pi)$
"cosine" = $a / \sqrt{(a + b)(a + c)}$
"faith" = $a + 0.5d / a + b + c + d$
"jaccard" = $a / a + b + c$
"phi" and "cor" = $ad - bc / \sqrt(R1 x R2 x C1 x C2)$
"rr" = $a / a + b + c + d$

Value

A symmetric similarity matrix

Author(s)

Alexander Christensen <[email protected]>

References

Choi, S. S., Cha, S. H., & Tappert, C. C. (2010). A survey of binary similarity and distance measures. Journal of Systemics, Cybernetics and Informatics, 8, 43-48.

Examples

# Simulate Datasets
one <- sim.fluency(10)

# Compute similarity matrix
cos <- similarity(one, method = "cosine")

# Simulate Datasets
one <- sim.fluency(10)

# Compute similarity matrix
cos <- similarity(one, method = "cosine")

Statistical tests for `bootSemNeT`

Description

Computes statistical tests for bootstrapped networks from bootSemNeT

Usage

test.bootSemNeT(
  ...,
  test = c("ANCOVA", "ANOVA", "t-test"),
  measures = c("ASPL", "CC", "Q"),
  formula = NULL,
  groups = NULL
)
test.bootSemNeT(
  ...,
  test = c("ANCOVA", "ANOVA", "t-test"),
  measures = c("ASPL", "CC", "Q"),
  formula = NULL,
  groups = NULL
)

Arguments

`...`	Object(s) from `bootSemNeT`
`test`	Character. Type of statistical test to be used. `"t-test"` Computes t-tests for all measures `"ANOVA"` Computes ANOVAs and includes Tukey's HSD for pairwise comparisons (`TukeyHSD`) `"ANCOVA"` Computes ANCOVAs that control for the number of nodes and edges in the networks and includes adjusted means and Tukey's HSD for pairwise comparisons (`TukeyHSD`)
`measures`	Character. Network measures to be tested. `ASPL` Average shortest path length `CC` Clustering coefficient `Q` Modularity statistic
`formula`	Character. A formula for specifying an ANOVA structure. The formula should have the predictor variable as "y" and include the names the variables are grouped by (e.g., `formula = "y ~ group_var1 * group_var2"`). See Two-way ANOVA example in examples
`groups`	Data frame. A data frame specifying the groups to be input into the formula. The column names should be the variable names of interest. The groups should be in the same order as the groups input into `bootSemNeT`

Value

Returns a list containing the objects:

`ASPL`	Test statistics for each proportion of nodes remaining for ASPL
`CC`	Test statistics for each proportion of nodes remaining for CC
`Q`	Test statistics for each proportion of nodes remaining for Q

If two groups:

A matrix in each object has the following columns:

`t-statistic`	Statistic from the `t.test`
`df`	Degrees of freedom
`p-value`	p-value with values equal to `0` being p < .001
`d`	Cohen's d
`CI95.lower`	Lower bound of the 95 percent confidence interval
`CI95.upper`	Upper bound of the 95 percent confidence interval
`Direction`	Direction of the effect. The argument `groups` will specify specifically which group is higher or lower on the measure. If no groups are input, then `"d"` and `"p"` are used to represent `data` and `paired` samples from `bootSemNeT`, respectively

Row names refer to the proportion of nodes remaining in bootstrapped networks

If three or more groups:

A list containing two objects:

`ANOVA`	A matrix containing the F-statistic, group degrees of freedom, residual degrees of freedom, p-value, and partial eta squared `p.eta.sq`
`HSD`	A matrix containing the differences between each group (`diff`), lower (`lwr`) and upper (`upr`) bounds of the 95% confidence interval, and the adjusted p-value (`p.adj`)

Author(s)

Alexander Christensen <[email protected]>

Examples

# Simulate Dataset
one <- sim.fluency(20)
two <- sim.fluency(20)

# Run partial bootstrap networks
two.result <- bootSemNeT(one, two, prop = .50, iter = 100,
sim = "cosine", cores = 2, type = "node", method = "TMFG")

# Compute tests
test.bootSemNeT(two.result)


# Two-way ANOVA example
## Simulated data
hihi <- sim.fluency(50, 500)
hilo <- sim.fluency(50, 500)
lohi <- sim.fluency(50, 500)
lolo <- sim.fluency(50, 500)

## Create groups
groups <- matrix(
c("high", "high",
"high", "low",
"low", "high",
"low", "low"
), ncol = 2, byrow = TRUE)

## Change column names (variable names)
colnames(groups) <- c("gf","caq")

## Run partial bootstrap networks
boot.fifty <- bootSemNeT(hihi, hilo, lohi, lolo, prop = .50,
type = "node", method = "TMFG", cores = 2, iter = 100)
boot.sixty <- bootSemNeT(hihi, hilo, lohi, lolo, prop = .60,
type = "node", method = "TMFG", cores = 2, iter = 100)

## Compute tests
test.bootSemNeT(boot.fifty, boot.sixty,
test = "ANOVA", formula = "y ~ gf*caq", groups = groups)


# Simulate Dataset
one <- sim.fluency(20)
two <- sim.fluency(20)

# Run partial bootstrap networks
two.result <- bootSemNeT(one, two, prop = .50, iter = 100,
sim = "cosine", cores = 2, type = "node", method = "TMFG")

# Compute tests
test.bootSemNeT(two.result)


# Two-way ANOVA example
## Simulated data
hihi <- sim.fluency(50, 500)
hilo <- sim.fluency(50, 500)
lohi <- sim.fluency(50, 500)
lolo <- sim.fluency(50, 500)

## Create groups
groups <- matrix(
c("high", "high",
"high", "low",
"low", "high",
"low", "low"
), ncol = 2, byrow = TRUE)

## Change column names (variable names)
colnames(groups) <- c("gf","caq")

## Run partial bootstrap networks
boot.fifty <- bootSemNeT(hihi, hilo, lohi, lolo, prop = .50,
type = "node", method = "TMFG", cores = 2, iter = 100)
boot.sixty <- bootSemNeT(hihi, hilo, lohi, lolo, prop = .60,
type = "node", method = "TMFG", cores = 2, iter = 100)

## Compute tests
test.bootSemNeT(boot.fifty, boot.sixty,
test = "ANOVA", formula = "y ~ gf*caq", groups = groups)

Triangulated Maximally Filtered Graph

Description

Applies the Triangulated Maximally Filtered Graph (TMFG) filtering method (Please see and cite Massara et al., 2016). The TMFG method uses a structural constraint that limits the number of zero-order correlations included in the network (3n - 6; where n is the number of variables). The TMFG algorithm begins by identifying four variables which have the largest sum of correlations to all other variables. Then, it iteratively adds each variable with the largest sum of three correlations to nodes already in the network until all variables have been added to the network. This structure can be associated with the inverse correlation matrix (i.e., precision matrix) to be turned into a GGM (i.e., partial correlation network) by using LoGo. See Details for more information on this network estimation method.

Usage

TMFG(data, depend = FALSE)
TMFG(data, depend = FALSE)

Arguments

`data`	Matrix or data frame. Must be a square matrix
`depend`	Boolean. Is network a dependency (or directed) network? Defaults to `FALSE`. Set to `TRUE` to generate a TMFG-filtered dependency network (output obtained from the `depend` function)

Details

The TMFG method applies a structural constraint on the network, which restrains the network to retain a certain number of edges (3n-6, where n is the number of nodes; Massara et al., 2016). The network is also composed of 3- and 4-node cliques (i.e., sets of connected nodes; a triangle and tetrahedron, respectively). The TMFG method constructs a network using zero-order correlations and the resulting network can be associated with the inverse covariance matrix (yielding a GGM; Barfuss, Massara, Di Matteo, & Aste, 2016). Notably, the TMFG can use any association measure and thus does not assume the data is multivariate normal.

Construction begins by forming a tetrahedron of the four nodes that have the highest sum of correlations that are greater than the average correlation in the correlation matrix. Next, the algorithm iteratively identifies the node that maximizes its sum of correlations to a connected set of three nodes (triangles) already included in the network and then adds that node to the network. The process is completed once every node is connected in the network. In this process, the network automatically generates what's called a planar network. A planar network is a network that could be drawn on a sphere with no edges crossing (often, however, the networks are depicted with edges crossing; Tumminello, Aste, Di Matteo, & Mantegna, 2005).

Value

Returns an TMFG filtered adjacency matrix

Author(s)

Alexander Christensen <[email protected]>

References

Christensen, A. P., Kenett, Y. N., Aste, T., Silvia, P. J., & Kwapil, T. R. (2018). Network structure of the Wisconsin Schizotypy Scales-Short Forms: Examining psychometric network filtering approaches. Behavior Research Methods, 50, 2531-2550.

Massara, G. P., Di Matteo, T., & Aste, T. (2016). Network filtering for big data: Triangulated maximally filtered graph. Journal of Complex Networks, 5, 161-178.

Examples

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(100), method = "cor"))

# Pearson's correlation only for CRAN checks
A <- TMFG(similarity(sim.fluency(100), method = "cor"))

Simulated Result for Dataset One and Two

Description

A result of bootSemNeT from two simulated datasets

Usage

data(two.result)
data(two.result)

Format

two.result (list, length = 6)

Examples

data("two.result")
data("two.result")

Plots for Vignette

Description

Plots for vignette taken from Christensen & Kenett (2019)

Usage

data(vignette.plots)
data(vignette.plots)

Format

vignette.plots (list, length = 3)

References

Christensen, A. P., & Kenett, Y. N. (2019) Semantic network analysis (SemNA): A tutorial on preprocessing, estimating, and analyzing semantic networks. PsyArXiv.

Examples

data("vignette.plots")
data("vignette.plots")

Package 'SemNeT'

Help Index

SemNeT–package

Description

Author(s)

References

See Also

Frequency of Animal Responses

Description

Usage

Format

Examples

Average Shortest Path Length

Description

Usage

Arguments

Value

Author(s)

References

Examples

Bootstrapped Semantic Network Analysis

Description

Usage

Arguments

Value

Author(s)

Examples

Clustering Coefficient

Description

Usage

Arguments

Value

Author(s)

References

Examples

Community Network Estimation

Description

Usage

Arguments

Value

Author(s)

References

Examples

Plots Networks for Comparison

Description

Usage

Arguments

Value

Author(s)

References

Examples

Convert Adjacency Matrix to Cytoscape Format

Description

Usage

Arguments

Value

Author(s)

References

Examples

Convert Network(s) to igraph's Format

Description

Usage

Arguments

Value

Author(s)

Examples

Equate Groups

Description

Usage

Arguments

Value

Author(s)

Examples

Finalize Response Matrix

Description

Usage

Arguments

Value

Author(s)

Examples