Cyber Times International Journal of Technology & Management

Growth of technology demands that more and more computational learning and pattern identification systems incorporate some form of machine learning, to facilitate learning from data and make predictions without much human intervention. One of the most recent approaches on the horizon is Deep Learning which is based on neural networks and attempts to capture high level abstractions of data. The core work in this domain has been concentrated on finding better representation models and also on creating models capable of working with large amount of unlabeled data. The main idea of this paper is to explore some of the prominent applications of Deep Learning like image identification, speech recognition, natural language processing etc. that have been successfully implemented and verified. The paper results in a novel means of classifying deep learning attempts that can provide insight to the applications of deep learning and could be a starting point to explore deep learning. The paper aims at presenting a new learner with a concise and clear glimpse into the pioneer work that has been done in the field till date. It can help in recognizing further possible avenues where deep learning could be applied.