Skip to content

SSSuperDan/gCore

Repository files navigation

gCore: Exploring Cross-layer Cohesiveness in Multi-layer graphs

Execution

Usage

gCore mode [graph_path graph_name] [options]

Mode:

cs      Run case study.
grk     Generate random \mathbf{k} vectors.
grp     Generate random \mathbf{p} vectors.
grkp    Generate random (\mathbf{k},\mathbf{p}) vector pairs.
pgcs	Compare the efficiency of CORE, dCC and GCS on pillar multi-layer graphs.
bp      Build P-tree.
bkp     Build KP-tree.
gcs     Compare the efficiency of CORE, dCC(pillar mlg only), RCD(general mlg only), GCS and (K)P-tree-based search (under different P-tree compaction levels).
gcii	Show statistics of GCI, including construction time, memory cost and number of nodes.
kv  	Compare the k-value of nodes in the (\mathbf{k},\mathbf{p})-core, the (\mathbf{k},\mathbf{ck})-rcd and the k-core.
pv  	Compare the p-value of nodes in the (\mathbf{k},\mathbf{p})-core, the (\mathbf{k},\mathbf{ck})-rcd and the k-core.
sm      Compute the size distribution when varying \mathbf{k}[i] and \mathbf{p}[i] for each layer i.
smk     Compute the size distribution when fixing \mathbf{k} and varying \mathbf{p}[i] for each layer i.
info	Compute basic information of the loaded multi-layer graph.

Options:

-ntc	Number of testcases.
-skf	File of sampled coreness vectors (\mathbf{k}).
-spf	File of sampled fraction vectors (\mathbf{p}).
-skpf	File of sampled (\mathbf{k}, \mathbf{p}) vector pairs.
-k      Coreness vector (\mathbf{k}).
-p      Neighbor coverage fraction vector (\mathbf{p}).
-b/pb   P-tree builder.
-s      Incremental step of coreness vectors to construct P-trees.
-o      Output path.
-ik     Start/Initial coreness vector.
-ck     Cross-layer degree threshold vector.
-pk     Coreness threshold for the selected layer.
-ptf	P-tree file.
-kptf	KP-tree file.
-f2if	Fraction to index map file.
-ps     Incremental step of fraction vectors to compute size matrices.
-ek     End coreness vector.
-d/dim  Dimension.
-g      Graph to perform case study, "dblp" and "twitter" are availiable.

P-tree builder options:

naive	Build P-tree without optimization.
ne      Build compact P+-tree with subtree elimination.
se      Build compact P+-DAG with subtree merge
nese	Build compact P+-DAG with both subtree elimination and subtree merge.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published