This project presents a clinically interpretable system for early detection of autism using machine learning integrated with explainable artificial intelligence (XAI). The main goal is to design a reliable and transparent classification model that can predict the likelihood of autism based on behavioral and developmental features. The system employs supervised learning algorithms such as K-Nearest Neighbors, Naive Bayes, Support Vector Classifier, and Random Forest Classifier to identify patterns within the dataset. To enhance interpretability, LIME (Local Interpretable Model-Agnostic Explanations) is implemented to provide clear explanations for each prediction. The dataset undergoes preprocessing, encoding, and normalization to ensure accurate learning and performance. A web-based application built using Flask enables user interaction through modules for registration, login, prediction, and logout. The project emphasizes transparency, accessibility, and interpretability, making the results understandable for users without deep technical expertise. The integration of XAI ensures that the decision-making process is traceable and trustworthy. This framework demonstrates that early detection of autism can be supported through interpretable machine learning approaches, promoting data-driven insights and human-understandable reasoning

Keywords: Autism, Machine Learning, Explainable AI, LIME, Classification, Interpretability, Flask, Early Detection, Transparency, Data Analysis