Structuring big data (big data) to train the CNN-ELM neural network in the tasks of programming recognition systems

Rasulov Mirzo Maksudovich

Based on the theory of scalability of machine learning, within the framework of this study, the concept of neural network training in the tasks of programming recognition systems is proposed. The large-scale approach has some limitations. In this study, to solve the problems of programming recognition systems, an approach is proposed for integrating asynchronous ELM components into the CNN convolutional network, based on the MapReduce parallel computing platform as a classifier module. This approach can save a lot of training time than single CNN-ELM models trained alone. This approach can improve scalability efficiency in machine learning systems by combining convolution and scaling approaches.

Keywords:big data structuring, deep learning, convolutional, neural network, big data.