R libraries for scientometrics on artificial intelligence in agricultural monitoring

Abstract

The deep integration of artificial intelligence (AI)-driven processes with adaptive crop monitoring design has emerged as a pivotal driving force for the environmental modelling of dynamic agriculture landscapes. AI-native mapping represents a fundamental change for cartographic solutions in engineering, natural and technical sciences, since it embeds automation into methodologies. This is especially important for Geographic Information Systems (GIS) where automation of spatial data processing is essential. Agricultural landscapes transform seasonally and yearly, which requires precise environmental forecasting. In this study, we provide an overview of recent methodological advancements in three interdisciplinary areas: environmental monitoring of agriculture landscape dynamics in soil studies, AI applications in GIS (machine learning (ML) and deep learning (DL) techniques), and bibliometric analysis using R based libraries (Bibliometrix, Treemap and Wordcloud), and Mendeley reference system. We investigate how novel AI and ML methodologies have been applied to scalable data-driven analysis in agriculture and soil studies and discuss the issues associated with their application. This review is based on the critical pool of over 100 papers indexed in the recognised databases Scopus, Web of Science (WoS), PubMed, and Google Scholar for in-depth analysis of AI applications in soil and environmetnal studies. We outline future perspectives for AI in environmental analysis, identifying best practices for AI implementation in GIS and systematic benchmarking in remote sensing for soil studies.