Paper Title
DEVELOPING AN AI-DRIVEN CHATBOT FOR SKIN DISEASE PREDICTION: CUSTOMIZED CNN AND DECISION TREE INTEGRATION

Abstract
Skin disease prediction chatbot project aims to create a chatbot that can help people predict the type of skin disease they may have based on their symptoms. The chatbot will utilizes a Customised Convolutional Neural Networks (CNN) model to classify user-uploaded skin images and create a Decision Tree model to predict the likely diagnosis based on the user's questionnaire responses, a customized CNN model will be employed. This CNN model will undergo training using a diverse dataset of skin images, enabling it to categorize these images into various classes, including squamous cell carcinoma, basal cell carcinoma, melanoma, nevus, actinic keratosis, and seborrheic keratosis. The CNN model will acquire the ability to discern unique patterns and features associated with each category, facilitating accurate predictions for new, unseen images.The chatbot will use a decision tree to ask a series of questions about the user's symptoms, such as the location of the rash, its colour, and any associated symptoms. The chatbot will then use this information to make a prediction based on the Decision Tree model with an accuracy 91.557%. The chatbot's interaction with the user will primarily rely on Natural Language Processing (NLP) techniques. Its purpose is to provide reply to general user questions. The primary aim of the project is to develop a user-friendly, accessible, and dependable tool that can assist individuals who might not have access to or the means to afford a dermatologist. This will be achieved by leveraging advanced machine learning techniques, including Convolutional Neural Networks (CNN) and Decision Trees, and incorporating NLP into the chatbot's functionality Keywords - CNN, Decision-Tree, NLP, SkinDisease Prediction, Images.