Generative Networks for Image Generation in East Asia’s Innovation Systems: A Bibliometric Analysis

Generative Networks commonly referred to as GANs or “Generative Adversarial Networks”, are a category of multi-layer perceptron that has emerged in recent years based on unsupervised type of machine learning. Since their conception, they have been widely studied all over the world due to their vast potential in multiple fields; image synthesis, computer vision, data augmentation and natural language processing, to name a few. It is an ongoing research topic with new network architectures being proposed for various purposes despite the existence of a variety of publications. Generative networks have been widely studied and the research in this field continues to evolve. This study aimed to provide an in-depth analysis of the present state of generative networks for image generation. The analysis is based on a thorough examination of publications, authors, funding sponsors and affiliated institutions from reputable databases like “Scopus” and “Web of Science”. Additionally, the study explored network characteristics such as co-authorship patterns, collaborations between countries, contributions, citations and keywords for a thorough analysis. This analysis revealed the substantial involvement of East Asian countries, notably China, within the realm of generative networks. The primary funding sponsors for this research predominantly hailed from East Asian countries, particularly China, South Korea and Japan. Additionally, the largest volume of documents and authors also originated from China.