Data Study Groups are week-long events at The Alan Turing Institute bringing together some of the country’s top talent from data science, artificial intelligence, and wider fields, to analyse real-world data science challenges. Using machine learning to improve sleep habits in Dementia patients Dementia is a collection of several types of cascading loss of brain functions that start with mild symptoms eventually leading to severe impairment which becomes terminal within a decade after diagnosis. In the UK, there are currently around 850,000 people living with dementia.There are several types of dementia, including Alzheimer’s disease, which accounts for 60-80% of all dementia cases [Ass22]. An estimated 80% dementia patients have sleep problems. Recent research has found sleep disturbances are a risk factor for deteriorating prognosis.Sleep disorders adversely impact the dementia patients’ physical, behavioural, and cognitive functions. Their sleeping problems often distress their caregivers as well, since patients with dementia are usually awake requiring care for uncomfortably long periods. Studies suggest that to improve sleep in dementia patients, pharmaceutical interventions are less effective.In comparison, non-pharmaceutical interventions have proved effective; light therapy has been successfully used in several pilot studies as means to regulate the circadian rest activity cycle. Designing such an intervention strategy is challenging due to the multi-dimensional nature of the problem. Multiple factors contribute to sleep problems, for example changes in the patient’s environment, their physical, cognitive and psychiatric conditions, neurodegenerative changes in the brain, medication used for chronic illnesses, and other dementia-related symptoms, among others.This challenge aims to predict the efficacies of the intervention markers and find the most effective ones which will allow the medical practitioners to take more informed suggestions.