Abstract In this paper we conducted a systematic literature study on image classification using transfer learning techniques during last five years using bibliometric methods. Transfer learning is an important method to classify images using existing neural network architectures. These architectures were developed using Convolutional Neural Networks concept of Deep Learning. The analysis is carried on the standard SCOPUS dataset and analyzed using VOSviewer software. The study is limited to publications in English language on “Transfer Learning”, “Deep Learning”, “Convolutional Neural Network” and “Image Classification” covering Computer and Engineering subjects during the years 2017 to 2021. This paper’s research will aid relevant researchers in understanding the current state of development and trends in this area. Only a few literature studies have tracked the growth of this field, and even fewer have used bibliometric approaches or scientific maps. As a result, this work provides an updated evaluation of this fast-growing subject, using a bibliometric technique to highlight new breakthroughs using scientific maps, and a unique visualization to depict the thematic network structure and progress.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|