Agent-based Social Simulation
Agent-based Social Simulation applied to Radicalism Spread
This project was developed in the Intelligent Systems Group within the Trivalent European project which aims to a better understanding of root causes of the phenomenon of violent radicalisation in Europe in order to develop appropriate countermeasures, ranging from early detection methodologies to techniques of counter-narrative. A full description of this project can be found on the digital repository of the UPM.
Agent-based Social Simulation tools have been developed as one of the applications of agent technology. It constitutes the intersection of three scientific fields, namely, agent-based computing, the social sciences, and computer simulation. Due to its increasingly use in social context, in this thesis, a web interface for its visualization is presented.
The aim of this project is to develop a web application which allows the user to visualize and analyze the results of a simulation. It allows the user to analyze data in real time and launch new simulations which can be configured from the web. For this purpose, a model which tries to simulate the growth of radicalism in a society is developed.
The model aims at improving the understanding of the influence of social links on radicalism spread. The model consists of two main entities, a spread model and a network model. The network model updates the agent relationships based on proximity and homophily, it simulates information diffusion and updates the agents’ beliefs. The model has been implemented in Python with the agent-based social simulator Soil and it has been evaluated using a sensitivity analysis while the application uses D3.js, which is a powerful JavaScript library, for rendering the results of the simulation and analyze them.
Tasio Méndez, J. Fernando Sánchez-Rada, Carlos A. Iglesias & Paul Cummings (2019). Analyzing Radicalism Spread Using Agent-Based Social Simulation. In Engineering Multi-Agent Systems (pp. 263–282). Springer International Publishing.
DOI: 10.1007/978-3-030-25693-7_14
This project has been accepted in the 6th International Workshop on Engineering Multi-Agent Systems which is hold in Stockholm, Sweden. EMAS 2018 aims to gather researchers and practitioners working in the domains of agent-oriented software engineering, programming multi-agent systems, declarative agent languages and technologies, machine learning, and AI in general to present and discuss their research and emerging results in MAS engineering.
