‘An automated data analytics platform for forensic testing’
My PhD project will use forensic data such as hair samples in order to create a data analytics tool that can be used to accurately predict whether a person has taken a specific drug or not. Currently, the outcome of substance use testing is determined by humans, where various factors are considered which lead to a decision. This procedure is of course subject to error and inconsistency.
The purpose of my project will be to use machine learning techniques to create a decision making tool that can accurately determine the outcome, therefore taking away the risk of incorrect judgement, error or inconsistency. The outlook is that this tool will be so precise that it can be used in the court of law, by judges, to aid with their final decision.
One example of when this may be used is in the case of family law, where a parent may have been accused of drug taking and if true could potentially lose custody of their child. If the wrong decision is made in such cases the effects can be detrimental to the family as the parent may wrongly lose their child, and a child loses their parent. On the other hand a child could be left in danger with a parent who is taking drugs but has been wrongly dismissed. The hope is that the tool I create can reduce or even eradicate these risks.
The partner organisation for this project is Forensic Testing Services. This project is funded by the Economic and Social Research Council (ESRC).
I graduated from University of Liverpool in 2016 with a bachelors degree in Psychology BSc. I then started at the University of Manchester in 2017 where I first completed a master's degree in Social Research Methods and Statistics MSc. I am currently enrolled on an integrated PhD course under the Data Analytics and Society Centre for Doctoral Training, also at the University of Manchester.