Harnessing Sustainable Development in Image Recognition Through No-Code AI Applications: A Comparative Analysis

• Published in: Recent Trends in Image Processing and Pattern Recognition Vol. 1576; pp. 146 - 155
• Main Authors: Kling, Nico, Runte, Chantal, Kabiraj, Sajal, Schumann, Christian-Andreas
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2022; Springer International Publishing
• Series: Communications in Computer and Information Science
• Subjects: Inequality; Labour market shift; No-code AI; SDGs; Sustainable development
• ISBN: 3031070046; 9783031070044
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-031-07005-1_14

Artificial intelligence (AI) solutions and sustainable development have increasingly received public attention and research interest. In this study, the authors discuss the emerging trend of no-code AI solutions, with regards to their contribution to achieving the Sustainable Development Goals, specifically goal 8 (decent work and economic growth) and goal 10 (reduced inequalities). To demonstrate the opportunities that no-code AI may facilitate, the authors compare the performance of conventionally coded models with a no-code model created in Microsoft Lobe, based on secondary data from a dataset offered by Kermany et al. [1] of chest x-rays for pneumonia detection. A total of 5840 JPEG images is used for training and testing, 1575 for normal and 4265 for pneumonia, respectively. Results indicate that the output generated by the studied no-code solution can keep up with coded ones, and partly even outperform them. Possible applications for industries and society include usability cases beyond image recognition, the application in citizen science as well as the exploration of economic development opportunities of no-code AI. Finally, no-code AI could perhaps offer an alternative and emerge as an industry best practice for delivering efficient low-cost solutions in emerging markets, where demographic data is scattered across various homogenous groups.