<![CDATA[Background: Greenwashing refers to misleading sustainability claims not backed by real actions, commonly seen in the oil and gas industry due to its dependence on fossil fuels. While companies may publicly commit to sustainability, their investments often contradict these claims, obstructing global renewable energy efforts. This mismatch between statements and actions misleads stakeholders and complicates audit processes. As demands for transparency grow, there is a pressing need for systematic tools to detect greenwashing. Prior research highlights that the narrative format of sustainability reports makes manual detection difficult, underscoring the need for technology-based solutions. Methods: This study aims to examine the application of Natural Language Processing (NLP), particularly the N-Gram model, in identifying indications of greenwashing in the oil and gas industry. The research uses a qualitative approach with a Systematic Literature Review (SLR) method and applies the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework. Findings: The N-Gram model aids in feature extraction by converting raw text from sustainability reports into structured representations and detecting linguistic patterns commonly found in overstated sustainability claims. When combined with classification methods like Support Vector Machine (SVM), it improves the accuracy of greenwashing detection. Key findings show that NLP can support auditors in assessing greenwashing risks and improving the efficiency of sustainability audits. Moreover, the integration of this technology promotes greater transparency in corporate disclosures. Conclusion: The application of the N-Gram model in the NLP context is effective in detecting greenwashing practices that were previously difficult to identify manually. Novelty/Originality of this article: This study offers novelty through the application of the N-Gram NLP model within the oil and gas industry context, which has been rarely explored in previous research. The practical implications of this study open opportunities for cross-sectoral implementation and the development of data-driven greenwashing identification standards in the future.]]>
Copyrights © 2025
