This is the source code for our ICDE2022 submission:

ID: 627, Title: Scalable Motif Counting for Large-scale Temporal Graphs.

Fast-temporal-motif is an algorithm proposed to count sub-graphs (i.e., motifs) in temporal graphs. This repository provides the source code of the algorithm. Besides, researchers can directly compile the project to have a quick start to verify the correctness and efficiency of the proposed algorithm.

1. download the project, the project looks like:

/FAST-temporal-motif
├── include
│   ├── global.h
│   ├── egeLoader.h
│   ├── starAndPairCount.h
│   ├── triangleCount.h
│   └── convertMartix.h
├── src
│   ├── global.cpp
│   ├── egeLoader.cpp
│   ├── starAndPairCount.cpp
│   ├── triangleCount.cpp
│   ├── convertMartix.cpp
│   └── main.cpp
├── makefile
└── README.md

2. execute make to compile the project, and get FAST_temporal_motif.exe

make

3. you can do make clean to remove FAST_temporal_motif.exe and all .o files

make clean

you can modify makefile to change compile settings. The default settings:

1. compile the project without using OPENMP. You can set the OMPFLAG to true to compile the project into a openmp program if you want it work with multiple threads
2. use -O3 Optimize

./FAST_temporal_motif -input -output -f -t -d -w

for example, if you want to count all temporal-motifs within the time span of 600 in input.txt, with 8 threads, thrd=200000 and output results to output.txt, you can run the program as follows:

./FAST_temporal_motif -input input.txt -output output.txt -f 1 -t 8 -d 200000 -w 600

The order of the arguments can be changed

n rows, each row represents a directed edge

3 data in each row, representing node 1, node 2, and timestamp

use space as separator

a 6*6 martix, each element representing the number of motif instances for $M_{i,j}$ in our paper.

Results of our experiments are shown in appendix.pdf