Many efforts are being made towards application of Machine Learning (ML) to solve Text Extraction problems, Text predictions to improve for betterment of society. In Educational institutions text extraction is widely practiced technique to examine inner elements of a document for clinical analysis and verification intervention, as well as for visual recognition of functioning of internal documents. This research project involves analysis of text from images for predicting text from documents available pu blicly. The research is subdivided into two major tasks, one is to build a model to understand usefulness of features in dataset, second to build a ML model for, then final task is to evaluate features extracted from processing document images for verification.