Abstract
Smart farming is a development that highlights the use of technologies such as the internet of things, cloud computing, machine learning, and artificial intelligence in the farm management cycle. For sustainable agriculture to adapt the ongoing change in climate and social structure is a major challenge for scientists and researchers. The approach needs information from various sources and its use in the relevant field, which lead to a growing interest in knowledge discovery from large data. Data mining techniques provide effective solutions for this problem as it supports the automation of extracting significant data to obtain knowledge and trends, the elimination of manual tasks, easier data extraction directly from electronic sources, and transfer to secure electronic system of documentation, which will increase the agriculture productions from same limited resources. In a nutshell, the aim of this chapter is to gain insight into the applications of data mining techniques in smart farming, which direction to employ sustainable agriculture and identify the challenges to be addressed.
TopIntroduction
Smart farming (SF) involves farming management concept using modern technology and the integration of information and communication technologies (ICT) into machinery, equipment, and sensors for use in agricultural production systems to increase the extent and value of agricultural products. The smart farming system is an autonomous & sophisticated mechanism, which will aid in the growth of agriculture yield by applying hi-tech agriculture techniques without human intervention. Agricultural practices are expected to advance the leading notion of “smart farming” (Tyagi, 2016; Babinet, Gilles et al., 2015) as it is being supported by biotechnology (Rahman, et al., 2013) and emerging digital technologies such as cloud computing (Hashem, et al., 2015), artificial intelligence (AI) & robots, Internet of Things (IoT) (Weber & Weber, 2010), and remote sensing (Bastiaanssen, et al., 2000).
Traditional farming practices are resource intensive in terms of capital, land, water, and fossil fuel use. But it threatens future food production by reducing biodiversity and contributing to environmental degradation and climate change which lower yields. Modern agriculture focuses on the high degree of mechanization, large scale, use science and technology, high capital and equipment, and use biotechnology to produce organic pollution-free food.
Farmers in the twenty-first century have access to Global Positioning System (GPS), soil scanning, data management, and IoT technologies. Agricultural enterprises have started collecting large amounts of data that includes soil and crop properties, which enables higher operational efficiency also. So, growth in this data size requires an automated method to extract necessary data.
Data mining is the process of finding correlations or patterns among the features in large relational databases. The overall goal of the data mining process is to extract information from a data set to extract previously unknown, interesting patterns such as groups of data records (cluster analysis), unusual records (anomaly detection) and transform it into an understandable structure for further use, which is used for all data mapping & processing. The technology of data mining is narrowly connected to data storage and is intertwined with the database management system. By applying data mining techniques, it is possible to extract useful knowledge, find patterns and trends in huge data.
Smart farming system covers main issues of farming such as water required for the crop at a particular stage, fertilizer to be used according to the micro-nutrients as well as macronutrients present in the soil, and the pesticides to be used depend on various environmental factors such as humidity, pollution (air, soil and water) etc.
Smart farming technologies require more and more professional skills to analyze agricultural data; extracting useful information has become the question of great importance. The implications of these situations highlight the need to consider the applications of data mining techniques to overcome the various problems faced by farmers in agriculture to grow good yield with SF for sustainable agriculture.
Since the growth in agriculture data is huge (Miller, McCarthy & Zakzeski, 2009), information is often hidden, so data mining techniques are used for their detection. Data mining techniques can be useful for finding the schemes of marketing in data, which are valuable and interesting for crop management (Mucherino & Rub, 2011). Both data mining and smart farming are relatively trending concepts, so it is expected to have knowledge about these applications and their implications for research and development be not widely spread.
Key Terms in this Chapter
Information and Communication Technologies (ICT): The technology that use telecommunications to provide access to the information.
Micronutrients: Nitrogen, Potassium, Phosphorous.
Cloud Computing (CC): Storing and accessing data and programs in the remote server over the Internet, rather than a local server or a personal computer.
Nearest Neighbor (NN): A non-parametric method used for regression and classification.
Macronutrients: Calcium, Magnesium, Sulphur.
Knowledge Discovery in Database (KDD): Refers to the broad process of discovering useful knowledge in data.
Rule-Based Classifiers: A sets of rules for classification and uses term absence and presence for the classification.
Decision Tree (DT): A graph illustrates every possible outcome of a decision using a branching method.
Fuzzy Logic: A computing method uses “degrees of truth” logic rather than the usual “true or false” Boolean logic.
Artificial Intelligence (AI): AI is a field of study which tries to make computers “smart” and the ability of a computer program or a machine to think and learn.
Genetic Algorithms: A heuristic search method used in artificial intelligence and computing.
Rough Sets: A new mathematical approach to imperfect knowledge and defined by two sets, the lower approximation and the upper approximation.
Internet of Things (IoT): A network used to collect and exchange data among the internet connected objects.
Support Vector Machine (SVM): It is a discriminative classifier formally defined by a separating hyperplane.
Artificial Neural Network (ANN): A computational model similar to the structure and functions of biological neural networks.
Application Programming Interfaces (APIs): The code with the protocols that govern the access point(s) for the server and allows applications to communicate with one another.
Global Positioning System (GPS): A satellite navigation system used to determine the ground position of an object.
Bayesian Networks (BN): A marked cyclic graph describes random variables and conditional dependencies.