Abstract
The article presents an introductory review of various approaches of deep learning including convolutional neural networks (CNNs), deep belief networks (DBNs), and auto-encoders (AEs). Each of these deep learning models is currently being used effectively in various fields such as medical application with healthcare systems, clinical trials, pharmacy industry, finance, agribusiness, energy industries, etc., and these models and all these models are extremely essential for any data scientist's toolbox. These deep learning models must build classes that should be flexibly designed, which can be useful in building new oriented application structure designs. Subsequently, for future development in the artificial intelligence-based technological world, it is important to have a necessary understanding of these deep learning models, which have been attempted to be refined through this systematic meta-analysis.
TopFocus Of The Article
This chapter focusing on systematic review of the deep learning strategies over various field like agriculture, medical, transportation etc. This review chapter is arranged in the following sections: section 2 covers Convolutional Neural Networks, section 3 discussed about Deep Belief Networks, finally section 4 covers Auto-Encoders. Overall discussion and challenges during preparation of learning model cover under section 5. At last, with conclusion this chapter completes in section 6.
Key Terms in this Chapter
Data Science: Data science is word derived from data and science which include statistics, scientific methods, artificial intelligence, data analysis in order to extract information from data.
Deep Learning (DL): Deep learning is subset of machine learning that is used to build learning models that can help to understand and analyze large data and support complex predictions for decision making.
Convolutional Neural Network: Convolutional Neural Network (CNN) is a form of feedforward neural network which make use of convolution, ReLU function, and pooling layers which help in dealing with information with high measurements.
Unsupervised Learning: Unsupervised learning make use of artificial intelligence algorithms for identifying pattern in data sets containing data points that are neither classified nor labeled.
Neural Networks: It is a set of algorithm that is used for recognizing the relationship between set of data by means of learning process.
Prediction Models: Prediction model predicts the future event or results by means of analyzing the pattern in given set of input data.
Supervised Learning: It is an approach of Artificial Intelligence where computer algorithm is trained on input data that has been labeled for a particular output.
Machine Learning (ML): Machine learning is a new technology where machine learns from the past data in order to decide and work for future data, thus helps in process of automation.