Due to the exponential growth of data being produced daily, automation is required in the education sector, where it is impossible for a single person to make sense of the data, even for a relatively straightforward task like question generation for a test. Producing Multiple Choice Questions (MCQs) is a necessary step in developing question pairs to measure student skills. In this research Natural Language Processing (NLP) and Deep Learning (DL) models are used for automatic generation of MCQs. An algorithm is introduced that can take information from text input and generate multiple-choice questions automatically. The words or sentences are chosen according to a predetermined pattern derived from the bag of words. The selected words or sentences and bag of words are provided to Convolution Neural Network (CNN) model for automatic MCQs generation. This research proposes an Intellectual Natural Language Processing Model with Bag of Words using CNN (INLP-BoW-CNN) for accurate generation of MCQs. The proposed model when contrasted with traditional models achieved better accuracy of 98.6%.