Using Machine Learning and the K-Nearest Neighbour Classification Method to Assess Autism Risk

The K-Nearest Neighbour (KNN) classification approach, one of the machine learning algorithms examined in this work, is used to determine a person's likelihood of developing autism spectrum disorder (ASD). The study use a dataset that encompasses a wide range of elements and practical training, such as behavioural characteristics, demographic data, and clinical markers linked to Autism Spectrum Disorder (ASD). The K-nearest neighbours (KNN) technique, renowned for its simplicity and efficacy in classification tasks, is used to examine patterns within the dataset and provide predictions pertaining to the probability of autism. The research starts with the first stage of data preparation, which access the processes of feature selection and normalisation. These steps are undertaken to optimise the performance of the model. Following this, a thorough assessment of the K-Nearest Neighbours (KNN) method is undertaken, taking into account various hyperparameter configurations and evaluating its efficacy in properly categorising and accessing people into Autism Spectrum Disorder (ASD) and non-ASD groups. The results demonstrate the effectiveness of the algorithm in identifying subtle patterns within the dataset, so enhancing our comprehensive comprehension of the elements that influence the risk of autism. The study findings have broader significance outside the field of diagnostics, as the model's precision that was established has the capacity to support clinicians and researchers in identifying and intervening early with people who are at risk of Autism Spectrum Disorder (ASD). This work makes a valuable contribution to the continuing endeavours aimed at improving the accuracy and effectiveness of autism risk assessment via the integration of machine learning approaches. Ultimately, this research aims to provide timely and targeted therapies, leading to enhanced results.