Abstract
Convolutional Neural Network (CNN) is an algorithm among deep learning. It is an effective method of solving common Natural Language Processing tasks such as text categorization. In this paper, we investigate the text categorization problem using a deep learning approach in particular CNN method. However, due to the rise of Big Data and increased complexity of tasks, the efficiency of CNNs have been severely impacted because of their extensive training time and high computational cost. To overcome these obstacles, we propose a MapReducebased CNN by reformulating the CNN's training from a single training network to a series of parallel trained smaller networks. Each smaller network processes a sub-sample of input text.