Deep learning to solve problems in various sector: A systematic literature review Nisa Aulia Saputra, Ida Hamidah, Agus Setiawan
Universitas Pendidikan Indonesia
Abstract
In the era of revolution industry 4.0, there are many problems in multiple sectors of life. The issues are challenging to solve, ranging from issues in the education quality performance system, manufacturing systems, construction, quality control, etc. Various efforts have been made to solve these problems, starting from the conventional method carried out by manually retrieving data to using a big data-based approach with deep learning. The success of deep learning in solving various problems proves that it provides very significant results. The purpose of writing this systematic review is to review the studies that have been carried out regarding the application of deep learning to solve problems that exist in various sectors. This systematic review shows an overview of deep learning neural networks created in the completion process. It shows the differences in the intelligent methods used, the advantages and disadvantages of deep learning in various models, and identifies future challenges and recommendations. The intelligent methods used in this systematic review are Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Artificial Neural Network (ANN), and Deep Neural Network (DNN) models. In ANN, the neural network is structured in the form of datasets of tables, images, and text. RNN is used to solve time series problems, text data, and audio data. DNN is used to decide an existing parameter or value. The methods used in this systematic review include search strategies, selecting literature, as well as managing and extracting data. The systematic review results concluded that CNN is the most widely used for this deep learning. That^s because CNN uses an algorithm and the image-based data transformation strategy for managing the data. The CNN deep learning model where the data is used can transform various 2D and 3D images with neural networks. Finally, deep learning has become very popular because it can transform various types of data to get the
Keywords: Deep Learning, Neural Network, Big Data, Convolutional Neural Network, Recurrent Neural Network, Artificial Neural Network, Deep Neural Network
