Early identification and diagnosis of brain tumors have been a difficult problem. Many approaches have been proposed using machine learning techniques and a recent study has explored deep learning techniques which are the subset of machine learning. In this analysis, Feature extraction techniques such as GLCM, Haralick, GLDM, and LBP are applied to the Brain tumor dataset to extract different features from MRI images. The features which have been extracted from the MRI brain tumor dataset are trained using classification algorithms such as SVM, Decision Tree, and Random Forest. Performances of traditional algorithms are analyzed using the accuracy metric and stated that LBP with SVM produces better classification accuracy of 84.95%. Brain tumor dataset is input to three-layer convolutional neural network and performance has been analyzed using accuracy which is of 93.10%. This study proves that CNN performs well over the machine learning algorithms considered in this work.

Green AI is an innovative project designed for sustainability development goals to aware people about climate change by decreasing carbon emissions across multiple domains. It consists of five core functionalities: garbage classification, travel optimization, electricity consumption analysis, device- based task management, and code optimization. This paper includes the methodologies, algorithms, and system overview of Green AI. Each category aims not only to minimize co2 emissions but also to provide alternative solutions to the user. From Machine learning models to user design, Green AI aims to provide best user experience moving towards sustainability. This project aims to make significant contributions towards a greener and more sustainable future by user actions.