rogerthatroach / datajam_luciana

DataJam Canada 2021 - KnowTech

This is a repository for research projects in the human trafficking domain. It includes all future related project ideas or research topics that would be useful for our project.

Each topic would have a separate repository. This would include the following:

1. Problem Statement

Human exploitation has been around no matter how far back in time we look. Currently, there has been more focus because of the interest shown by organizations like the United Nations which has shown the global scale of the problem and the enormous impact on the young and the migrant communities. When families are having to choose between food for the rest of the family and the trafficking of a child, the solution is not going to be identifying the number of cases in the past but rather in building solutions which help identify the perpetrators and the circumstances under which the human trafficking flourishes.

Human Trafficking Example (Forced Labour / Servitude)

Many people in South Asia migrate for work to the Middle East countries of Saudi Arabia , Kuwait and Qatar, etc. where there is a booming Construction and Oil and Gas business. The workers are recruited by company Agents who operate in the feeder countries. The basic pitch to the migrant worker is that Accommodation, Food, a living wage, healthcare, etc. will be provided. The salary quoted is multiple time what the person would expect to get in their home country. The hidden catch is the Agent Fee and Transportation to and from the hosting country. The amounts are usually so high that the migrant person is forced into debt which on paper the person would be able to pay off within a few months of working in the Middle East.

The Reality that is faced by the migrant worker is much different. Upon landing or very soon after, the host will take the passport, pay much lower salary, there would be no accommodation except for a rotating bed in a room with 8 other persons. The trap is now even more difficult to escape from because not only does the migrant worker not have enough money to live the in the country but his debt is not decreasing.

It is a truly horrific situation for the migrant worker.

2. Objective

The objective of building a solution to the human trafficking problem was to build a set of tools which can proactively identify (a) the perpetrator, (b) the victim, both potential and currently a victim and, (c) the environment under which the human trafficking event is about to happen or did happen. The Datasets provided for the DataJam gave in idea of the scope of human trafficking and the number of countries. We found nearly 50 incidents identified as being related to human trafficking. The events were reported from nearly 80 countries including the UNHCR. The charts in Appendix 1 shows the scope of the human trafficking incidents and where these incidents have occurred.

3. Solution/Data use case description

1. Identify Malicious Tweets
2. Classify Handles into known Profiles
3. Warn potential victims about predatory tactics based upon keywords used by Perpetrators, writing style, etc.

4. Pitch

A pitch of maximum 4 minutes.

Hint: Recommended is to add here just link(s), for example to Youtube recording:

Example location: link to video

GitHub Limitations:

• 100MB per file
• 10GB per repository

5. Datasets

Location: /datasets

A list of data sets, sources and challenges for the research project. This would include:

• API access to required TAH data
• Access to other datasets (either through API or downloaded)
• Statements on data quality issues, transformations needed
• Container for new data generated for this research (i.e. through merging other data sets)

6. Project Code

Location: /project

• NLP Model (EDA, Transformations, Visualisation, BERT, Prediction)
• Twitter Data Generator using Twitter API
• User Interface code

7. Additional docs (Optional)

Location: /docs

• PowerPoint presentation
• Flayers
• Additional videos/demo
• Protocols
• Guides