In the rapidly evolving domains of e-commerce and artificial intelligence (AI), this thesis explores the integration of AI to automate product description generation for ecommerce platforms. Given the significant contributions of e-commerce to global sales and the transformative potential of AI in marketing strategies, this research addresses the critical need for efficient and scalable content creation amidst the expanding range of online products. Employing the Design Science Research Methodology (DSRM), the study develops and integrates an OpenAI-based content generator capable of producing product descriptions. The effectiveness of this generator is evaluated through controlled experiments that evaluate AI-generated content by humans and ChatGPT itself, focusing on ethical compliance, adherence to specified constraints, and linguistic quality. The findings demonstrate that AI can automate the generation and evaluation of product descriptions with high accuracy in ethical compliance and constraint adherence, though linguistic errors, particularly inflection errors, were noted. Additionally, the AI model showed potential in automating the evaluation of product descriptions, albeit with some limitations compared to human evaluators. This advancement not only contributes to the fields of computer science and AI, but also offers practical solutions to contemporary challenges in e-commerce, potentially improving the consumer shopping experience and enabling smaller businesses to compete more effectively online. Keywords: E-commerce, Artificial Intelligence, Content Generation, Design Science Research Methodology, Marketing
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:lnu-130080 |
Date | January 2024 |
Creators | Mierkhan, Shirin, Åkesson, Caroline |
Publisher | Linnéuniversitetet, Institutionen för datavetenskap och medieteknik (DM) |
Source Sets | DiVA Archive at Upsalla University |
Language | English |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.1385 seconds