Modern Slavery is defined by the Modern Slavery Act 2015 as a person holding another in slavery, servitude or requires them to perform forced or compulsory labour. Human trafficking, also an offence under the Act, is defined as the recruitment, movement, harbouring or receiving of children, women or men through the use of force, coercion, abuse of vulnerability, deception or other means for the purpose of exploitation.
Survivors of modern slavery can be of any nationality, gender and age but children, women and vulnerable adults are often targeted. Their needs vary, shaped by their experiences, characteristics, and personal aspirations. Our understanding of survivors’ needs are constantly evolving and stakeholders need to adapt support services to ensure they fulfil legal obligations and keep pace with what the survivors actually need.
RESTART will support survivors in having their voice heard and provide evidence for policy makers seeking to develop existing strategies and programmes. How? We will analyse case notes from engagements with survivors in order to discover what have they asked for to-date. This will be done using a custom-built suite of natural language processing tools to analysing large volumes of survivor statements and stories, pulling out key insights that would take human analysts an incredibly long time to achieve – if at all. In addition, an app will be created to ask direct questions and obtain further data on survivors needs and aspirations.
Trilateral Research will develop NLP models and visualise insights in its inhouse built Honeycomb dashboard. For more on Honeycomb see here.
RESTART aims to provide policy makers with a better understanding of survivor needs, how they should be prioritised and how these change over time. It helps survivors by aligning support packages with what they actually need. It is our hope that this project will change the way stakeholders see the potential of data science methods – in particular NLP- in engaging with survivors, giving them a voice and providing them with support.