RANKING OF ARTIFICIAL INTELLIGENCE TOOLS TO ASSIST IN THE WRITING OF SCIENTIFIC ARTICLES USING THE GAUSSIAN AHP METHOD

Abstract

Due to the complexity in the preparation of articles for publication in scientific journals, there is a growing use of artificial intelligence (AI) tools to support researchers and increase their productivity and efficiency. This article aims to rank AI models programmed to assist the researcher in the scientific writing process. The criteria evaluated are related to the research problem, the state of the art, the gap and the bibliographic reference. The methodology consisted of scoring the AI tools on a scale of 0 to 10, where zero would be an indication of no potential and ten of high potential of the AI tool. And also, rank the AIs through the multicriteria decision-making method Gaussian AHP of Operations Research, having as alternatives ChatGPT3.5, Elicit, Dimensions AI, Paper Digest, Semantic Scholars, Connected Paper, Jenni.ai, SciSpace Copilot and Grok. Results showed that Semantic Scholars was identified as the most effective alternative, evidencing the importance of the "Bibliographic Reference" criterion. The lower influence of the "Research Problem" highlights the greater predominance of other factors in the choice of these tools, such as the state of the art and the gap. These results offer insights for researchers, highlighting the need for context-specific considerations and continuous improvements in the integration of AI into the academic environment.