Use and Adaptations of Machine Learning in Big Data—Applications in Real Cases in Agriculture

Abstract

The data generated in modern agricultural operations are provided by diverse elements, which allow a better understanding of the dynamic conditions of the crop, soil and climate, which indicates that these processes will be increasingly data-driven. Big Data and Machine Learning (ML) have emerged as high-performance computing technologies to create new opportunities to unravel, quantify and understand agricultural processes through data. However, there are many challenges to achieve the integration of these technologies. It implies making some adaptations to ML for using it with Big Data. These adaptations must consider the increasing volume of data, its variety and the transmission speed issues. This paper provides information on the use of Big Data and ML for agriculture, identifying challenges, adaptations and the design of architectures for these systems. We conducted a Systematic Literature Review (SLR), which allowed us to analyze 34 real cases applied in agriculture. This review may be of interest to computer or data scientists and electronic or software engineers. The results show that manipulating large volumes of data is no longer a challenge due to Cloud technologies. There are still challenges regarding (1) processing speed due to little control of the data in its different stages, raw, semi-processed and processed data (value data); (2) information visualization systems, which support technical data little understood by farmers.

Machine Learning in Agriculture: A Comprehensive Updated Review

“…Furthermore, other studies were concerned with the implementation of ML methods regarding the main grain crops by investigating different aspects including quality and disease detection [ 29 ]. Finally, focus has been paid on big data analysis using ML, aiming at finding out real-life problems that originated from smart farming [ 30 ], or dealing with methods to analyze hyperspectral and multispectral data [ 31 ].…”

“…For this reason, more user-friendly systems should be developed. In particular, simple systems, being easy to understand and operate, would be valuable, as for example a visualization tool with a user-friendly interface for the correct presentation and manipulation of data [ 25 , 30 , 31 ]. Taking into account that farmers are getting more and more familiar with smartphones, specific smartphone applications have been proposed as a possible solution to address the above challenge [ 15 , 16 , 21 ].…”

Artificial Neural Networks and Computer Vision’s-Based Phytoindication Systems for Variable Rate Irrigation Improving

“…The work [28] provides an overview of the use of machine learning in agriculture, namely, the management of water resources, soil, and animal husbandry. Big data and machine learning (ML) open up new possibilities for understanding agricultural processes using data, along with this, there are numerous problems of integrating these technologies that require research and solution [29]. Machine learning is an important decision support tool, for example in [30] it has found application in yield forecasting, including making decisions about crop types and cultivation technologies during the growing season.…”

Evaluation of Random Forests (RF) for Regional and Local-Scale Wheat Yield Prediction in Southeast Australia

“…These involve building tree ensembles whose growth are controlled by randomized selection of (input-output) vectors from the training dataset; which are then assembled as classification or regression models to predict the most likely output class (or values) from the inputs of the test dataset with good accuracy and robustness to outliers with lower likelihood of generalization errors [ 15 ]. RF have the potential to generate better models compared to single decision-tree models [ 16 ], are more efficient computationally and therefore suitable for regional and global applications in agriculture [ 17 ] where Big Data dominates [ 18 , 19 ]. For instance, RF-driven yield prediction for sugar cane in Australia has been found to be more accurate and reliable than traditional approaches such as multiple linear regression [ 20 , 21 ].…”