Abstract
Autism is a progress state linked with healthcare expenses, therefore, primary transmission of autism signs can reduce on these budgets. The autism transmission method involves offering a sequence of problems for parents, caregivers, and family members to reply on behalf of the kid to conclude the possible of autistic characters. Often present autism transmission tools, for instance the Autism Quotient (AQ), comprise various questions, furthermore to alert design of the questions, which makes the autism broadcast process prolonged. One would-be solution to advance the proficiency and correctness of transmission is the reworking of ambiguous rule in data mining. Ambiguous rules can be taken out robotically from earlier controls and cases to form a transmission cataloging system. This system can then be applied to predict whether individuals have any autistic behaviors rather than trusting on the unadventurous domain expert rules. This paper assesses fuzzy rule-based data mining for predicting autistic signs of kids to speak the above-mentioned problem. Empirical results show high performance of the ambiguous data mining model in respect to analytical correctness and sensitivity rates and unexpectedly lower than projected specificity rates when compared with new rule-based data mining models.