Agricultural Prediction Using Hybrid Butterfly Optimization with Convolutional Neural Network

The climate changes will have a considerable impact on agriculture and food security invariably of the region and crop. Agriculture urgently has to adapt to the expanding global population in order to ensure future food security for this population. As the need of the food is increasing globally via quality and in terms of quantity, there is a need of industrialized support to field of agriculture. A group of cutting-edge technologies known as “Internet of Things” (IoT) have enormous potential to modernize the agriculture industry. Modern technology uses the term Precision agriculture in this era. A vital part of soil and water conservation on agricultural land is irrigation. An effective platform for predicting soil moisture content in the future utilizing current soil and environmental factors may be available for agricultural land irrigation needs. A significant amount of data is gathered in agriculture, and several DM techniques are employed to effectively utilize it. A number of methods related to data mining techniques are covered in this work. The Soul of Infinite Life, often known as the soil, is what keeps life on earth alive. Agriculture continues to be India’s biggest employer and source of income despite great improvements in the service sector. In order to estimate yield, this research introduces a hybrid ML model using IoT. Using a CNN-based recursive architecture, this extracted data has been categorized. Comparing the proposed system of agriculture prediction to the current system of DT, LR, SVM, and WSVM, which provides poor accuracy, sensitivity, specificity, and precision, as well as an F1-score, HBOCNN techniques are used to get the desired results. In the fields of cultivating crops, predicting the weather, managing wildlife, forestry, raising animals, identifying markets, and funding rural areas, the study cites examples of IoTs that have helped these communities' agricultural requirements.