Neural Transfer Learning in NLP for Post-Traumatic-Stress-Disorder Assessment

This is the first article of a sequel with different dimensions of my experience in the Omdena PTSD Challenge (fine-tuning ULMFit for our use case, ML Rest Backends with MLFLow, baseline text classifier, etc.). For the last seven weeks, I have been contributing as a volunteer Lead Machine Learning Engineer in this AI for Good challenge, coordinating tasks and leading a team of more than 30 ML Engineers in unknown charters of Machine Learning and Deep Learning. The main goal of the project was to research and prototype technology and techniques suitable to create an intelligent chatbot to mitigate/assess PTSD in low resource settings. My general contribution is to evaluate UMLFit as a possible solution avenue to tackle this hard and impactful problem. "The challenge is to build a chatbot where a user can answer some questions and the system will guide the person with a number of therapy and advice options."