Data Science for Agricultural Innovation and Productivity

One of the disruptive technologies that will emerge in the 21st century is the digital twin, which is a digital copy of any physical object that may exist in any setting. Many industries heavily rely on digital twin technology to produce high-quality products that can be shipped throughout the world with no loss in efficiency. The initial efforts have been made by the agricultural sector toward the implementation of digital twin technology in farming and other types of activities. It has already begun to apply vertical farming together with other crucial cutting-edge technologies in a chosen number of smart cities. 

In the Indian economy, agriculture is one of the main economic resources. Unemployment is one of the biggest problems. In many areas, agriculture offers about 60% of jobs. Various roles and responsibilities are involved in the agricultural sector including farmers, farm equipment operators, technical specialists, and more. There are a number of opportunities available in the agricultural sector. In developing countries, unemployment has decreased significantly since agriculture accounts for 20% of GDP, and agriculture can yield much greater benefits. Crops are classified into commercial and non-commercial crops. Most people can access and use non-commercial crops in their daily lives. Tomatoes are considered non-commercial crop. The edible berry of the species, Solanum lycopersicum, is commonly known as the tomato plant. Tomato plants are affected by many different diseases and because of these diseases, the losses are heavy. The disease can affect leaf, stem, or tomatoes. Initially, the disease is observed in the leaves, and eventually the disease worsens. Anticipating disease at an early stage is a major concern and therefore preventive measures are taken to obtain good crop quantity and quality. Applying deep learning techniques to disease detection has many advantages over other machine learning techniques. Deep learning is a part of machine learning techniques that helps train computers to do things that are natural to humans. 

The agriculture sector not only contributes to the nation's economy but also serves as an important element in foreign exchange and trade markets. With the advancement in technology, robots, drones, satellite imagining, IoT, wireless sensor networks, machine learning, big data analytics, and unmanned aerial vehicles (UAV) are being deployed to manage, monitor and control agricultural chores. However, the farmers are unable to meet the increasing urban food demand with limited cultivable land availability. Thus, to solve this issue, hydroponic farming is opted for in several parts of the world. It is a soil-free and nutrient-rich water medium for agriculture, which is increasingly opted for by the urban population. Hydroponic farming has been vastly explored in the context of urban farming, where land, water, time, and labour are required in a limited amount, yet productivity is far better compared to traditional agricultural methods.

 It has been recently adopted in urban sections in India due to restricted movement in COVID-19 pandemic situations to fulfil basic food requirements. However, hydroponic farming has shortcomings such as higher initial cost, the possibility of complex nutrient discharge problems, the energy requirement for the creation of microclimatic conditions, fertigation and effluent treatment and pretrained skilled labour. In order to resolve these issues, a smart hydroponic farming architecture is discussed, which reduces human intervention and water wastage using wireless sensor networks and IoT. In order to successfully and efficiently implement the agricultural supply chain, machine learning algorithms and data mining techniques are utilized from the production to inventory storage stage. The following sections deal with a brief introduction to hydroponic farming, its architecture and components, and future opportunities regarding the field of automated hydroponic farming. 

In an agriculture-based society, where sustainable farming operations are required, quantitative field status and plant-by-plant monitoring may benefit all cultivators by enhancing farmland management. Sensing technology, artificial intelligence, autonomous robotics, and computerized data analytics will be important. In this book chapter, the essential features of using robotics in agriculture are presented; namely, the primary reasons for the automation of agriculture, the role of robotics in agriculture, its classification, evolution, and consideration of autonomous navigation for commercial agricultural robots, currently existing models of agriculture robots and their comparison, the potential benefits and limitations of agriculture robotics, gathering of massive data and using data science approaches for improving the food productivity and its influence on boosting the agriculture industry. The current study focuses on the adoption of agriculture robotics in the farming sector for various purposes (from land preparation to harvesting). The application of agriculture robotics for food production can favor the incorporation of agricultural robotics companies to minimize labor costs and food shortages. Furthermore, agriculture robotics can be the catalyst for new sources of information on the environmental impact (agroecological footprint) of the local food production chain.

War, pollution, and a plethora of other threats are prevailing in the world on a daily basis. The globe, which has a population of more than 7.3 billion, is constantly harmed by human activity. Climate change is one of the world's most lethal problems because of these severe deformations. UN reported that 10000 people have died in extreme weather events like fire and floods in the past two years. Globally, $280 billion is lost to climate catastrophes from 2021 to 2022. While a single action won't be able to stop or slow down climate change, many tiny contributions from several professions will help it have an emotional effect. Scientists from all over the world are looking for ways to manage the transformation of the landscape in order to anticipate the dangers of climate change and, if possible, to reduce their effects on the future of the earth. Particularly with the Internet of Things (IoT), it is possible to slow the increase in global temperatures and cut back on hothouse emigration. The Internet of Things (IoT) encompasses more than just cutting-edge gadgets and intelligent machinery since it affects the state of the planet, from its climate to its financial resources. If we use technology effectively, it may be the instrument that saves the world. Since it involves everything from monitoring ozone levels in a meat packing facility to keeping an eye on public trees for banks, environmental monitoring is a broad activity for the Internet of Things (IoT). These findings mark the beginning of the process of developing several interconnected architectures that will support cutting-edge services and have greater effectiveness and flexibility. 

Plant disease mechanization in the agricultural discipline is a major source of concern for every country, since the world's population continues to grow at an alarming rate, increasing the need for food. However, due to a scarcity of necessary infrastructure in various parts of the world, it is difficult to identify them quickly in some areas. In the context of the expanded use of technology, it is now feasible to assess the efficiency and accuracy of methods for identifying illnesses in plants and animals. It has recently been discovered that information technology-based tools, technologies, and applications are effective and realistic measures for the improvement of the whole agricultural field, spanning from scientific research to farmer assistance. The integration of expert systems as a strong tool for stakeholders in agricultural production has enormous promise, and it is now being explored. The suggested effort begins with the collection of disease symptoms and environmental factors by agriculture specialists and plant pathologists, who will then analyze the information gathered. The corrective solution is then recommended to the end user by an expert system, which is accessed through a mobile application. Computer application consisting of an expertise base, inference engine, and a user interface is envisaged as the machine of the future. Integrated inside the gadget is a structured expertise base that contains information on the signs and treatments of various ailments. In order to identify and diagnose plant disorders, the machine must first locate and diagnose the condition. It is accomplished by the analysis of the symptoms of illness on the crop's surface. On the basis of the yield and the surrounding environment, this symptom is utilized to identify the illness and give an entirely unique diagnostic solution. The computer will test the plants and their disordered lives inside the database and provide a set of diagnostic levels in accordance with the condition that the plants are suffering from, according to the database. Farmers may easily identify and manipulate plant diseases with the help of the suggested technology, which is supported by a sophisticated expert system.

The industry with the fastest global growth is aquaculture, particularly in developing nations. Studies of control systems, automation, IoT, and artificial intelligence are becoming more and more prevalent as a result of developments in information technology and embedded systems in the digitalization era. The 4.0 industrial revolution, which promotes device connection via IoT and AI, is a result of the swift advancement of science and technology. One of the key factors in the growth of this revolution is IoT. The fourth industrial revolution is a shift from manual to automated operations that rely on computer technology. The digitalization of all industries, including fisheries and aquaculture, must be the first step in technological advancement toward the usage of the Internet of Things (IoT) and Big Data. The raising and catching of fish have undergone several changes as a result of digitization in the fishing industry. The main benefits of digitizing fisheries are those related to time, money, and labor efficiency. The procedure of digitizing aquaculture can be seen from a variety of angles, including pre-production (locating cultivation sites using GIS), production (improving resource efficiency by employing automatic feeders and water quality testing equipment), and post-production (utilization of digital marketing). 

Darjeeling tea (Camellia sinensis) is famous worldwide for its excellent aroma and taste, and out of that, the best quality tea is produced in the Kurseong hill area. A year-long analysis of soil samples collected from different sites was done by applying the plate-count method in terms of colony-forming units for determining the presence of microbial population within soils cultivating tea on a monthly basis. Coordination in specific microbes might also be responsible for the impact. Fungi, bacteria, and actinomycetes – these three groups of microbes were tested during the process. Results showed that the neo tea plant, rhizosphere and rhizosphere of several other perpetual plants of various ages, flourishing in age-old tea gardens, seemed to expedite the growth of microbes. At present, the tea rhizosphere has been tested thoroughly, specifically in relation to plant-microbe response. Counter to the common outcomes, rhizosphere and soil ratios were found to be regularly less than 1 in samples collected from age-old tea gardens, showing an overall -ve rhizosphere impact. The finding of the 'negative rhizosphere effect' in old tea bushes is a significant and novel nature of the tea rhizosphere. The -ve impact on the rhizosphere of aged tea bushes does not seem to be a regular phenomenon that is related to the aging of plants generally but might be distinctive particular to tea plants. Other important and associated features include colonization of tea, rhizosphere, soil pH, etc. Supremacy of a certain population of microbes, an affinity towards a section of general opponents constitutes a good instance of reciprocated selection in the natural environment. These discoveries have unlocked newer paths for extended research in the field of 'rhizosphere microbiology'. The present study is an attempt to evaluate the transforming features coupled with the microbial activity and diversity in the tea rhizosphere and significant implications in the tea industry. 

The introduction of sensing-based technology has transformed the agriculture sector in many ways. This chapter explores the potential of sensing-based technology, including big data and artificial intelligence, in agriculture to lower production costs and increase yield efficiencies. The application of various sensors is explained in various sectors of agriculture, like crop farming, animal farming, and fish farming. This technology has the potential to automate farming and has the ability to shift to precise cultivation for higher crop yields and better quality while using the minimum resources. This chapter also elaborates on the different types of sensors used in agriculture, their benefits, and related issues for their various applications. These technologies have some real issues in the application, which need to be sorted out, and more efforts should be made to make the product more cost effective, relevant, and customized for the use of farmers.

A collection of best practices for publishing as well as connecting structured data on the Web is referred to as “Linked Data.” In the past ten years, the practice of Linked Open Data has significantly increased the structured data available on the Web. This structured data provides machine-readable descriptions of real-world items, opening hitherto unheard-of opportunities for research around natural language processing. With the aid of technology, the agriculture sector advanced significantly, becoming more data-driven and intelligent. Studies on how these data can be used, for what kinds of tasks, and to what degree they can be helpful for agricultural tasks, however, are lacking. Our paper puts an effort into providing an overview of concepts and technologies in the Semantic Web and their applicability to farming.