The NETWORK Procedure

Functional Summary

Table 2 summarizes the statements and options available in the NETWORK procedure.

Table 2: Functional Summary of Statements and Options

Description Option
PROC NETWORK Statement
Input
Specifies the in-memory graph to use GRAPH=
Specifies the links data table LINKS=
Specifies the links data table to query LINKSQUERY=
Specifies the nodes data table NODES=
Specifies the nodes data table to query NODESQUERY=
Specifies the nodes subset data table NODESSUBSET=
Output
Specifies the output data table to contain summary information about in-memory graphs OUTGRAPHLIST=
Specifies the links output data table OUTLINKS=
Specifies the nodes output data table OUTNODES=
Options
Specifies whether to enforce determinism DETERMINISTIC=
Specifies the graph direction DIRECTION=
Specifies whether to use a distributed graph DISTRIBUTED=
Specifies the index offset for identifiers INDEXOFFSET=
Specifies the desired frequency (in number of seconds) between log entries LOGFREQTIME=
Specifies the overall log level LOGLEVEL=
Specifies whether to include multilinks MULTILINKS=
Specifies the maximum number of threads to use for multithreaded processing NTHREADS=
Specifies whether to include self-links SELFLINKS=
Specifies that the input graph data are in a standardized format STANDARDIZEDLABELS
Requests that the output graph data include standardized format STANDARDIZEDLABELSOUT
Specifies whether time units are in CPU time or real time TIMETYPE=
Data Statements
LOADGRAPH Statement
UNLOADGRAPH Statement
Variables Statements
LINKSQUERYVAR Statement
Specifies the data variable name for the query path expansion lower bounds EXPANDLOWER=
Specifies the data variable name for the query path expansion upper bounds EXPANDUPPER=
Specifies the data variable name for the from nodes in the query graph FROM=
Specifies the data variable name for the to nodes in the query graph TO=
Specifies the data variables for the link attributes to read VARS=
Specifies the data variables for the link attributes to match in the query graph VARSMATCH=
LINKSVAR Statement
Specifies the data variable name for the auxiliary link weights AUXWEIGHT=
Specifies the data variable name for the from nodes FROM=
Specifies the data variable name for the to nodes TO=
Specifies the data variables for the additional link attributes to read VARS=
Specifies the data variables for the additional link attributes to write to the output results VARSOUT=
Specifies the data variable name for the link weights WEIGHT=
NODESQUERYVAR Statement
Specifies the data variable name for the nodes in the query graph NODE=
Specifies the data variables for the node attributes to read VARS=
Specifies the data variables for the node attributes to match in the query graph VARSMATCH=
NODESSUBSETVAR Statement
Specifies the data variable name for the nodes NODE=
Specifies the data variable name for the reach identifier REACH=
Specifies the data variable name for the sink indicator SINK=
Specifies the data variable name for the source indicator SOURCE=
NODESVAR Statement
Specifies the data variable name for the nodes NODE=
Specifies the data variables for the additional node attributes to read VARS=
Specifies the data variables for the additional node attributes to write to the output results VARSOUT=
Specifies the data variable name for the node weights WEIGHT=
Algorithm Statements
BICONNECTEDCOMPONENTS Statement
Specifies the output data table for biconnected components OUT=
Specifies the output data table for the links of the block-cut tree OUTBCTREELINKS=
Specifies the output data table for the nodes of the block-cut tree OUTBCTREENODES=
CENTRALITY Statement
Specifies how to calculate authority centrality AUTH=
Specifies how to calculate betweenness centrality BETWEEN=
Specifies whether to normalize the betweenness calculation BETWEENNORM=
Specifies how to calculate closeness centrality CLOSE=
Specifies a method for accounting for the shortest path distance between two nodes when a path does not exist (disconnected nodes) CLOSENOPATH=
Calculates the node clustering coefficient CLUSTERINGCOEFFICIENT
Specifies how to calculate degree centrality DEGREE=
Specifies how to calculate eigenvector centrality EIGEN=
Specifies the algorithm to use for eigenvector calculation EIGENALGORITHM=
Specifies the maximum number of iterations for eigenvector calculation EIGENMAXITERS=
Specifies how to calculate hub centrality HUB=
Specifies how to calculate influence centrality INFLUENCE=
Specifies how to calculate PageRank centrality metric PAGERANK=
Specifies the damping factor for the PageRank algorithm PAGERANKALPHA=
Specifies the convergence tolerance for the PageRank algorithm PAGERANKTOLERANCE=
Specifies the percentage of source nodes to sample for the approximate betweenness calculation SAMPLEPERCENT=
CLIQUE Statement
Specifies whether to calculate the clique number CLIQUENUMBER=
Specifies the maximum number of cliques to return during clique enumeration MAXCLIQUES=
Specifies the maximum link weight for the cliques found MAXLINKWEIGHT=
Specifies the maximum node weight for the cliques found MAXNODEWEIGHT=
Specifies the maximum size for the cliques found MAXSIZE=
Specifies the maximum amount of time to spend finding cliques MAXTIME=
Specifies the minimum link weight for the cliques found MINLINKWEIGHT=
Specifies the minimum node weight for the cliques found MINNODEWEIGHT=
Specifies the minimum size for the cliques found MINSIZE=
Specifies the output data table for cliques OUT=
COMMUNITY Statement
Specifies the community detection algorithm ALGORITHM=
Specifies which data variable defines groups of nodes to fix together in a community FIX=
Specifies the percentage of small-weight links to be removed LINKREMOVALRATIO=
Specifies the maximum number of iterations for community detection MAXITERS=
Specifies the output data table for intercommunity links OUTCOMMLINKS=
Specifies the output data table for the community summary OUTCOMMUNITY=
Specifies the output data table for the community level summary OUTLEVEL=
Specifies the output data table for the community overlap OUTOVERLAP=
Specifies the random factor in the parallel label propagation algorithm RANDOMFACTOR=
Specifies the random seed for the parallel label propagation algorithm RANDOMSEED=
Applies the recursive option to break large communities RECURSIVE
Specifies the resolution list for community detection RESOLUTIONLIST=
Specifies the modularity tolerance value for community detection TOLERANCE=
Specifies which data variable defines the initial node partition for warm starting community detection WARMSTART=
CONNECTEDCOMPONENTS Statement
Specifies the algorithm to use for connected components ALGORITHM=
Specifies the output data table for connected components OUT=
CORE Statement
Specifies the maximum amount of time to spend calculating the core decomposition MAXTIME=
CYCLE Statement
Specifies the algorithm to use for cycle enumeration ALGORITHM=
Specifies the maximum number of cycles to return during cycle enumeration MAXCYCLES=
Specifies the maximum length for the cycles found MAXLENGTH=
Specifies the maximum link weight for the cycles found MAXLINKWEIGHT=
Specifies the maximum node weight for the cycles found MAXNODEWEIGHT=
Specifies the maximum amount of time to spend finding cycles MAXTIME=
Specifies the minimum length for the cycles found MINLENGTH=
Specifies the minimum link weight for the cycles found MINLINKWEIGHT=
Specifies the minimum node weight for the cycles found MINNODEWEIGHT=
Specifies the output data table for the links of the cycles OUTCYCLESLINKS=
Specifies the output data table for the nodes of the cycles OUTCYCLESNODES=
NODESIMILARITY Statement
Specifies whether to calculate Adamic-Adar similarity ADAMICADAR=
Specifies the number of lowest-ranked node pairs to output BOTTOMK=
Specifies whether to calculate common neighbors similarity COMMONNEIGHBORS=
Specifies the convergence threshold for vector similarity CONVERGENCETHRESHOLD=
Specifies whether to calculate cosine similarity COSINE=
Specifies whether to calculate vector embeddings EMBED=
Specifies the variable names of precomputed vector embeddings EMBEDDINGS=
Specifies whether to calculate Jaccard similarity JACCARD=
Specifies the maximum similarity score to output MAXSCORE=
Specifies the minimum similarity score to output MINSCORE=
Specifies the number of dimensions for vector embeddings NDIMENSIONS=
Specifies the number of negative samples per positive sample used in training the model NEGATIVESAMPLEFACTOR=
Specifies the number of training samples for the vector algorithm NSAMPLES=
Specifies the similarity measure to use for node pair ranking ORDERBY=
Specifies the output data table to contain the convergence curves for vector similarity OUTCONVERGENCE=
Specifies the output data table to contain the similarity scores between pairs of nodes OUTSIMILARITY=
Specifies the proximity order for the vector similarity algorithm PROXIMITYORDER=
Specifies the sink node for similarity calculations SINK=
Specifies the source node for similarity calculations SOURCE=
Specifies whether to output sparse similarity data SPARSE=
Specifies the number of highest-ranked node pairs to output TOPK=
Specifies whether to calculate vector similarity VECTOR=
PATH Statement
Specifies the maximum length for the paths found MAXLENGTH=
Specifies the maximum link weight for the paths found MAXLINKWEIGHT=
Specifies the maximum node weight for the paths found MAXNODEWEIGHT=
Specifies the maximum amount of time to spend finding paths MAXTIME=
Specifies the minimum length for the paths found MINLENGTH=
Specifies the minimum link weight for the paths found MINLINKWEIGHT=
Specifies the minimum node weight for the paths found MINNODEWEIGHT=
Specifies the output data table for path links OUTPATHSLINKS=
Specifies the output data table for path nodes OUTPATHSNODES=
Specifies the sink node for path calculations SINK=
Specifies the source node for path calculations SOURCE=
PATTERNMATCH Statement
Specifies the code that defines FCMP functions CODE=
Specifies whether to filter matches by using the induced subgraph of the match INDUCED=
Specifies the FCMP function for link filters LINKFILTER=
Specifies the FCMP function for link-pair filters LINKPAIRFILTER=
Specifies the FCMP function for a filter that is based on a potential match (that is, any subset of nodes or links or both) MATCHFILTER=
Specifies the maximum number of matches to return MAXMATCHES=
Specifies the maximum amount of time to spend in pattern matching MAXTIME=
Specifies the FCMP function for node filters NODEFILTER=
Specifies the FCMP function for node-pair filters NODEPAIRFILTER=
Specifies the links output data table for the induced subgraph of matches OUTMATCHGRAPHLINKS=
Specifies the nodes output data table for the induced subgraph of matches OUTMATCHGRAPHNODES=
Specifies the links output data table for matching subgraphs OUTMATCHLINKS=
Specifies the nodes output data table for matching subgraph mappings OUTMATCHNODES=
Specifies the links query output data table to contain the generated queries OUTQUERYLINKS=
Specifies the nodes query output data table to contain the generated queries OUTQUERYNODES=
Specifies the summary output data table for pattern match queries OUTSUMMARY=
Specifies the data variable name for the query key QUERYKEY=
Specifies whether to aggregate nodes across query keys for creating the induced subgraph(s) QUERYKEYAGGREGATE=
PROJECTION Statement
Specifies whether to calculate Adamic-Adar similarity ADAMICADAR=
Specifies whether to calculate common neighbors similarity COMMONNEIGHBORS=
Specifies whether to calculate cosine similarity COSINE=
Specifies the projection method for directed graphs DIRECTEDMETHOD=
Specifies whether to calculate Jaccard similarity JACCARD=
Specifies the output data table to contain the lists of common neighbors between pairs of nodes OUTNEIGHBORSLIST=
Specifies the output data table to contain the links of the projected graph OUTPROJECTIONLINKS=
Specifies the output data table to contain the nodes of the projected graph OUTPROJECTIONNODES=
Specifies the data variable name for the partition flag PARTITION=
REACH Statement
Calculates the directed reach counts DIGRAPH
Treats each node as a source in reach calculations EACHSOURCE
Specifies the maximum number of hops in the reach calculations MAXREACH=
Specifies the output data table for reach counts OUTCOUNTS=
Specifies the output data table for reach links OUTREACHLINKS=
Specifies the output data table for reach nodes OUTREACHNODES=
SHORTESTPATH Statement
Specifies the maximum absolute objective gap for the paths found MAXABSOBJGAP=
Specifies the maximum link weight for the paths found MAXLINKWEIGHT=
Specifies the maximum number of ranked paths to find for each pair MAXPATHSPERPAIR=
Specifies the maximum relative objective gap for the paths found MAXRELOBJGAP=
Specifies the minimum link weight for the paths found MINLINKWEIGHT=
Specifies the output data table for shortest paths links OUTPATHSLINKS=
Specifies the output data table for shortest paths nodes OUTPATHSNODES=
Specifies the output data table for shortest path descriptive statistics OUTSUMMARY=
Specifies the output data table for shortest path weights OUTWEIGHTS=
Specifies which data variable in the nodes subset data table defines the sequence of nodes to visit SEQUENCE=
Specifies the sink node for shortest path calculations SINK=
Specifies the source node for shortest path calculations SOURCE=
SUMMARY Statement
Calculates information about biconnected components BICONNECTEDCOMPONENTS
Calculates information about clustering coefficients CLUSTERINGCOEFFICIENT
Calculates information about connected components CONNECTEDCOMPONENTS
Calculates the approximate diameter and chooses the weight type DIAMETERAPPROX=
Includes only finite values when calculating descriptive statistics that are related to shortest paths FINITEPATH
Specifies the output data table for summary results OUT=
Calculates information about shortest paths and chooses the weight type SHORTESTPATH=
TRANSITIVECLOSURE Statement
Specifies the output data table for transitive closure results OUT=


Table 3 lists the supported DIRECTION= option values in the PROC NETWORK statement.

Table 3: Supported Input Formats by Statement

Statement and Options UNDIRECTED DIRECTED
BICONNECTEDCOMPONENTS X
CENTRALITY
AUTH= X
HUB= X
 Otherwise X X
CLIQUE X
COMMUNITY X X
CONNECTEDCOMPONENTS
ALGORITHM=DFS X X
 Otherwise X
CORE X X
CYCLE X X
NODESIMILARITY X X
PATH X X
PATTERNMATCH X X
PROJECTION X X
REACH X X
SHORTESTPATH X X
SUMMARY
BICONNECTEDCOMPONENTS X
DIAMETERAPPROX= X
 Otherwise X X
TOPOLOGICALSORT X
TRANSITIVECLOSURE X X


For each algorithm statement in the NETWORK procedure, Table 4 indicates which output data table options you can specify and whether the algorithm populates the data tables that are specified in the OUTNODES= and OUTLINKS= options in the PROC NETWORK statement.


Last updated: November 22, 2022