Leveraging machine learning for advancing insect pest control: A bibliometric analysis

Abstract

Insects have flourished in various ecosystems owing to their evolutionary prowess. However, certain behaviours have led specific species to be classified as pests in human-dominated settings. Ensuring accurate pest identification and assessing risks are vital for both agricultural productivity and effective pest control. While traditional methods, based on manual checks and expert opinions, tend to be time-consuming and error-prone, machine learning (ML)—a branch of artificial intelligence—has brought groundbreaking shifts in computer vision and predictive analytics, paving the way for advanced agricultural methods. This study delves into a bibliometric analysis of the confluence between ML and pest control from 1999 to 2022. Drawing data from 2348 publications in the Web of Science (WoS) databases, we identified a marked uptick in interest after 2017—a decade marked by a 40-fold growth in publication numbers. An examination of 706 WoS core articles offered insights into temporal and geographic trends, co-citation patterns, key publications, and recurring keywords. Also, we spotlight major ML techniques employed in pest management and hint at promising directions for subsequent research. Overall, this paper serves as an exhaustive resource for individuals intrigued by the intersection of computer science and agriculture.