A Machine Learning Platform for NLP in Big Data

Abstract

The objective of this work is the design of an architecture for the management and storage of data that are exponentially increasing and coming from different sources. The main areas of intervention are focused on demonstrating how a solution of a service architecture is able to abstract itself from any software that produces or receives data and guarantees a correct flow of them regardless of the speed or quantity of input, also guaranteeing fault-tolerance. A case study is presented for processing and archiving natural language, ensuring fault tolerance, reliability, speed and a high storage capacity. The result of this paper is an architecture that is able to process text data and to obtain the subject from each of them.