modern-slavery-statements-research 0.1.2

A collection of tools to conduct research on the Modern Slavery Statements text corpus.

Research on Modern Slavery

This repository is going to contain a collection of experiments and analyses performed on the Modern Slavery Statements Dataset.

Introduction

The UN Sustainable Development Goal 8.7 states: Take immediate and effective measures to eradicate forced labour, end modern slavery and human trafficking and secure the prohibition and elimination of the worst forms of child labour, including recruitment and use of child soldiers, and by 2025 end child labour in all its forms.

In 2018, the Global Slavery Index found that there were 40.3 M people in modern slavery, of whom 25M were in forced labor producing computers, clothing, agricultural products, raw materials, etc and 15M were in forced marriage.

The Future Society, an independent nonprofit think-and-do tank launched a partnership with the Walk Free Initiative to automate the analysis of modern slavery statements produced by businesses to boost compliance and help combat and eradicate modern slavery. The team at The Future Society is curating an up-to-date repository of >16K modern slavery statements (and counting) to boost machine learning research in this area. The data is scraped based on the collection of report links provided by the modernslaveryregistry.org.

By sharing your analysis and contributing to this repository you help the global community to hold multi-national corporations accountable for how they treat their workforce and suppliers.

Prerequisites

• Python 3.6+ installed on your system
• If you'd like to use the provided tutorials, you also need access to a Jupyter notebook

Quickstart

It's recommended that you use a virtual environment such as virtualenv, pipenv or similar.

Option 1 - notebook

Copy this notebook and follow the tutorial.

The linked notebook above also shows you how to plot the word count distribution of all the documents downloaded.

Option 2 - command line

Install the package:

pip install modern-slavery-statements-research

Specify your AWS access credentials as -i (aws access key id) and -a (secret access key) arguments and run (without the curly brackets):

download-statements -i {aws_access_key_id} -a {aws secret access key}

The logs printed in the console will tell you the name of the data folder.

If you've set up your modern slavery project related AWS CLI credentials as default you can simply run

download-statements

You can explore more options by running download-statements --help

How to get data access?

The data is available in the modern-slavery-dataset-txt bucket in AWS S3. The future plans involve releasing the dataset for the general public access.

The bucket contains multiple copies of statements from various scraping runs so the logic is to always pick the latest folder. The provided scripts and examples take care of this automatically. The raw documents are found in the /data/raw_statements/ folder in the modern-slavery-dataset-raw bucket. Metadata is found in s3://modern-slavery-dataset-raw/data/ms_registry/ following the same logic.

As it's work in progress, at present, if you'd like to work with this data, please send an email to [e-mail] with a link to your social profile (linkedin, facebook or similar ) and you'll receive IAM user credentials on the first possible instance that would allow you to download and access the data.

Get Help

If you'd like to get help with domain expertise or technical requirements and implementations then get in touch with Adriana or Edgar respectively.

Roadmap

Over the next few weeks and months, the following improvements are planned to the dataset and the repository:

1. Provide a convenient one-command entry point to the data, including download to a pandas dataframe.
2. Improve the dataset quality by continuously including more documents and improving the data cleaning pipeline.
3. Provide examples of analysis.
4. Provide manually annotaded labels for a subset of the corpus to enable analyses using supervised methods.
5. Open source the data and research for public access.

Citation

If you intend to share any form of public research and analysis based on the data from this repository and the modern-slavery-dataset-raw and modern-slavery-dataset-txt buckets in AWS S3, then please include the following citation to your publication:

The Future Society. (2020) Modern Slavery Statements Research. Retrieved from https://github.com/the-future-society/modern-slavery-statements-research.

Contributions

If you'd like to contribute to the research then take a look at any of the issues or get in touch with Adriana or Edgar.

Take a look at colab notebooks based on the modern slavery corpus:

• Rey Farhan's initial text data exploration and assumptions' check.
• Parth Shah's exploration of knowledge graphs based on subject-object syntactic relations