Scholarlink

<p>Artificial intelligence (AI) is increasingly being used today to provide solutions to many environmental problems. This is because of AI’s capability over empirical measurement and conventional computational techniques. This work therefore aim to review publications on&#xa0;AI applications&#xa0;in environmental research&#xa0;field. Literature was search for in Scopus database using Boolean-keyword operators. Among others, the study objectives include identifying evolution of the publications on AI within environmental research&#xa0;domain; their research themes and trends. Mixed-methods (qualitative and quantitative analysis) was adopted for the study, and they&#xa0;comprise bibliometric, text-mining and content analysis. 797 suitable publications were included in the study. Result shows 83% of the included publications in our dataset were published between 2018 and 2024. By checking through authors’ keywords, it appears that recent and popular research interest is being paid to areas such as industry 4.0, smart cities, thermal comfort, circular economy, environmental monitoring, carbon emissions and others. Text-mining of the dataset shows that most frequently used ML technique is Artificial Neural network (ANN), followed by Support Vector Machine (SVM). Text-mining also identified three major themes from the publications on AI within environmental research. They are (i) ecological decision support system (DSS), for detection, prediction and analysis of changes in ecological aspects for evidence-based decisions; (ii) sustainability transitions, illustrated by circular economy, industry 4.0, sustainable supply chains, and (iii) pollution monitoring/-***- and treatment. Eleven major research themes were also identified from bibliographic coupling of the publications within the dataset. The study results could provide a basis for future areas of interests within environmental research.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial intelligence in environmental research: bibliometric, text mining and content analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年6月23日

查看全文
  • Chukwuebuka C. Okafor,
  • Festus A. Otunomo,
  • Valentine E. Nnadi,
  • Chinelo A. Nzekwe,
  • Adaobi V. Nwoye,
  • Charles C. Ajaero

摘要

Artificial intelligence (AI) is increasingly being used today to provide solutions to many environmental problems. This is because of AI’s capability over empirical measurement and conventional computational techniques. This work therefore aim to review publications on AI applications in environmental research field. Literature was search for in Scopus database using Boolean-keyword operators. Among others, the study objectives include identifying evolution of the publications on AI within environmental research domain; their research themes and trends. Mixed-methods (qualitative and quantitative analysis) was adopted for the study, and they comprise bibliometric, text-mining and content analysis. 797 suitable publications were included in the study. Result shows 83% of the included publications in our dataset were published between 2018 and 2024. By checking through authors’ keywords, it appears that recent and popular research interest is being paid to areas such as industry 4.0, smart cities, thermal comfort, circular economy, environmental monitoring, carbon emissions and others. Text-mining of the dataset shows that most frequently used ML technique is Artificial Neural network (ANN), followed by Support Vector Machine (SVM). Text-mining also identified three major themes from the publications on AI within environmental research. They are (i) ecological decision support system (DSS), for detection, prediction and analysis of changes in ecological aspects for evidence-based decisions; (ii) sustainability transitions, illustrated by circular economy, industry 4.0, sustainable supply chains, and (iii) pollution monitoring/-***- and treatment. Eleven major research themes were also identified from bibliographic coupling of the publications within the dataset. The study results could provide a basis for future areas of interests within environmental research.