1. Dr. SHAMBHU KUMAR SINGH - Assistant Professor, School of Computer Science and Engineering, Sandip University, Madhubani, Bihar,
India.
2. Dr. PARMANAND PRABHAT - Assistant Professor, School of Computer Science and Engineering, Sandip University, Madhubani,
Madhubani, Bihar, India.
India's groundwork is its husbandry. With over 60% of the workforce deputed and producing extinguished 18% of the heathen GDP, it is a on the map sector of the Indian finance. Although there are many driveway in which we can use processing to increase crop out-turn, a swain can only avail if he is able to sell his prolificacy. Three laws have been departed by the Indian kingship to encourage the exportation of predial crop athwart the nation. But today, we saakshi swain all extinguished the nation fighting against these precept to intercede their sound. Swain worry that big dealer will exploit them as stalking horse and lower the price at which they sell their commodities. After doing a thorough organic analysis of the place, we exalted the memory of creating a predial produce archest that facilitates direct intelligence between swain and retailers, allows for product commentary and crop yielding momentum prediction, and vaticinator the tariff of predial produce based on quantity produced and trailing years' sales local cuss. manmouji rains, unexpected temperature decreases, and sultriness waves have all been brought on by the transferable clime, and the ecosystem has suffered significant harm. Thankfully, doohickey learning has produced useful device for tackling international puzzle, such as husbandry. These climate change- concerned agricultural figure can be resolved by using manifold machine Educate methods. The purpose of this cantle is to create a method to identify screen diseases and suggest screens. For both objectives, publicly accessible datasets were spend. Regarding the gnaw recommendation system, portent extraction was done, and a maker of machine Educate methods were used to puff-puff the dataset, including Support Vector doohickey (SVM), Random Forest, conclusion Tree, Logistic Regression, and Multilayer Perception. 99.30% accuracy was attain via the random forest multiple. CNN architectures such as ResNet50, and EfficientNetV2 were trained and juxtapose for the bane distemper identification organization. EfficientNetV2 outperformed the convenience, with a high exactitude of 96.08%.
Groundwork, Finance, Memory, Tree, Organization, Convenience.