Associative classification mining in the behavior study of Autism Spectrum Disorder

The number of children diagnosed with Autism Spectrum Disorder (ASD) has increased in the past few years and the root cause of the symptom cannot yet be determined. The diagnosis today relies heavily on the observation of children´s behaviors. This paper presents a technique to investigate the behavior factor associations, and to classify these relations using classification based on association (CBA). Our experiments used actual patient profiles from two hospitals in Thailand. This dataset was categorized by doctors into two types: Autism and Pervasive Developmental Disorder - Not Otherwise Specified (PDD-NOS). Our analysis results show several interesting behavior patterns in autism disorder. These results provide valuable information for doctors to conduct further studies in the early intervention of autistic symptoms. The goal of our research is to develop a data analysis tool to aid doctors in the diagnosis process in the future.