Demystifying the Applications of Deep Learning Using Graph Learning Models and Algorithms

EasyChair Preprint no. 9922

Abstract

Machine learning algorithms that handle a variety of data kinds are typically the backbone of real-world intelligent systems. Due to its inherent complexity, graph data have presented machine learning with hitherto unheard-of hurdles despite their widespread use. Since graph data are embedded in an irregular domain, unlike text, audio, and image data, some fundamental operations of current machine learning methods cannot be used. To address these issues, numerous graph learning models and algorithms have been created. In this article, the most recent graph learning techniques are reviewed in detail, along with some of their possible uses. The article has several functions. For researchers and practitioners in a variety of fields, including social computing, information retrieval, computer vision, bioinformatics, economics, and e-commerce, it serves as a rapid reference to graph learning. Second, it offers details about active fields of study in the area. Thirdly, it hopes to pique interest in graph learning and inspire fresh research ideas.

Keyphrases: deep learning, graph signal processing, graph topology, network representation learning

@Booklet{EasyChair:9922,
  author = {A S Sasipriya},
  title = {Demystifying the Applications of Deep Learning Using Graph Learning Models and Algorithms},
  howpublished = {EasyChair Preprint no. 9922},

  year = {EasyChair, 2023}}