Smart farming using artificial intelligence: A review

• Published in: Engineering applications of artificial intelligence Vol. 120; p. 105899
• Main Authors: Akkem, Yaganteeswarudu, Biswas, Saroj Kumar, Varanasi, Aruna
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.04.2023
• Subjects: Deep learning; Machine learning; Smart farming; Time series analysis; Deep learning; Smart farming; Time series analysis; Machine learning
• ISSN: 0952-1976; 1873-6769
• DOI: 10.1016/j.engappai.2023.105899

Smart farming with artificial intelligence provides an efficient solution to today’s agricultural sustainability challenges. Machine learning, Deep learning, and time series analysis are essential in smart farming. Crop selection, crop yield prediction, soil compatibility classification, water management, and many other processes are involved in agriculture. Machine learning algorithms are used for crop selection and management, Deep learning techniques are used for crop selection and forecasting crop production, and time series analysis is used for demand forecasting of crops, commodity price prediction, and crop yield production forecasting. Crops are chosen using machine learning algorithms and deep learning algorithms based on soil, soil compatibility classification, and other factors. In the agriculture industry, this article offers a thorough review of machine learning and deep learning techniques. Crop data sets can be used to classify soil fertility, crop selection, and many other aspects using machine learning algorithms. Deep learning algorithms can be applied to farming data to do time series analysis and crop selection. Because there is more need for food due to the growing population, crop production forecasting is one of the crucial tasks. Therefore, future crop production must be predicted in order to overcome food insufficiency. In this article, several time series algorithms were reviewed. Suggesting appropriate crop recommendations using machine and deep learning by estimating crop yield by using time series analysis will reduce food insufficiency in the future. •Different technologies such as machine learning, deep learning, and time series analysis have been presented.•The study evaluated ensemble methods like stacking, advanced machine learning methods like Inducing Readable Oblique Decision Tree, SVM with Relief-F feature, KNN algorithm based on KD-Tree for crop selection.•Different Deep learning techniques like extreme learning machine, radial basis functional network networks related to crop selection has been discussed.•Smart farming with advanced machine learning techniques would provide better crop selection which will reduce farmer suicides and advanced machine learning techniques are more robust and fast learning speed.•Time series analysis algorithms reviewed and presented comparison.