2023-01-05: Using Social Media to Identifying Deceptive Advertisements to Counter Human Smuggling Project Funded by CCI

We are delighted to receive funding to work on a research project to develop new ways to counter human smuggling by identifying deceptive posts on social media sites. This is a joint work collaboration between Dr. Faryaneh Poursardar, Dr. Vikas Ashok from ODU CS, and Dr. Erika Frydenlund from VMASC and will support one graduate student for one year

Faryaneh Poursardar, Vikas Ashok, and Erika Frydenlund, From Gap to Gain: Counter Human Smuggling by Identifying Deceptive Advertisements in Social Media, Commonwealth Cyber Initiative (CCI) - Securing Interaction between Humans and Machines, 12/2022 – 12/2023, $60,000

Human smuggling across borders is one of the major global concerns affecting many countries across the world, including the United States. Research shows that smugglers are heavily relying on social media to solicit migrants attempting to cross into the US through deceptive advertisements promising safe and guaranteed means of crossing the border. Deceptive travel advertisement types have distinct visual and textual patterns that subtly alert people to the true nature of their “travel” services. This research focuses on an important yet under-explored topic in online misinformation and disinformation, namely countering the unscrupulous social media use by human traffickers and coyotes for the purpose of human smuggling.

This project fosters research in novel multi-lingual natural language processing (NLP) techniques for contextual embedding and interpretation of non-English texts in posts. Most existing NLP tools and techniques primarily cater to English language inputs. In this project, machine learning models will be trained that learn how to exploit the patterns to distinguish deceptive advertisements from legitimate ones. Also, no public dataset exists that covers deceptive smuggling advertisements in multiple languages from different countries across the globe and building such a dataset is another contribution to this research.

The Commonwealth Cyber Initiative (CCI) is establishing Virginia as a global center of excellence at the intersection of security, autonomous systems, and intelligence. With a mission of research, innovation, and workforce development, CCI serves as a catalyst for the commonwealth's economic development and long-term leadership in this sector.

One of the research themes selected by CCI is securing the interactions between humans and machines. Recent evolution in artificial intelligence, cyber-physical systems, and communications are leading to a world in which humans and autonomous machines increasingly interact. CCI views cybersecurity as intrinsically cross-disciplinary and devise solutions that lead to secure, resilient, and harmonious interactions between people and robots, drones, autonomous vehicles, and other cyber-physical systems. This approach let researchers explore opportunities for innovation (commercialization, entrepreneurship, etc.) in security and privacy for the interaction between humans and machines.

For this call, CCI funded 15 proposals. The breakdown by the four CCI nodes is:

  • NoVa – Northern Virginia: 5 (1 funded by CCI and 4 funded by the NOVA node)
  • CoVa – Costal Virginia: 2
  • CVN – Central Virginia Node: 3
  • SWVA – Southwest Virginia: 5 (2 funded by CCI and 3 funded by the Southwest node)

We are excited to be one of two funded projects from CCI for the COVA node and looking forward to working on this research collaboration.

--Faryaneh Poursardar (@Faryane)