Code necessary to reproduce the experimentation presented in A Multi-Objective Genetic Algorithm for Detecting Dynamic Communities using a Local Search driven Immigrant’s Scheme

Requirements

1. Docker and Docker-Compose
2. Having port 8888 free

Getting started

1. Open a terminal on the root folder of the proyect.
2. Run export USER_ID=id -u.
3. Run docker-compose run -d.
4. Chek that both containers, db and notebook, are running using docker-compose ps.
5. Open a browser and go to localhost:8888
6. A login page will ask you for a token. back on the terminal run docker-compose logs notebook | grep token.
7. Copy the token (token=XXXXXX...) from the terminal and use it to login at the browser.
8. Use the jupyter notebook inside host_data to reproduce the experiment.
9. Outputs will be generate inside tables and figures folders.

Structure

├── data
├── figures
├── Section 4.2 - reparation techniques selection.ipynb
├── sources
│   ├── experiment
│   ├── gas
│   ├── gloaders
│   ├── mongo_connection
│   └── reparators
├── start-notebook.sh
└── tables

• data: contains all the datasets used in the article's experimentation.
• figures: In this folders the figures shown in the article are generated.
• Section 4.2 - reparation techniques selection.ipynb: jupyter notebook to reproduce the article's experimnents
• sources: code of the different algorithms proposed.
• start-notebook.sh: a simple script that starts the jupyter notebook (used by docker-compose)
• tables: Folder where the article's tables are generated.

Sources structure

• experiment: all the logic necessary to run, plot and load reasults of an experiment. An experiment is composed of a dataset and MOGA settings.
• gas: all the logic neccesary to implement the different MOGA proposed in the article. Each MOGA has a config and an execution file.
• gloaders: contains all the logic neccesary to load, analize, and store a dataset.
• mongo_connections: contains all the logic neccesary to connect with the mongoDB database.
• reparators: contains all the logic neccesary to apply the local search operators.