Agriculture is one of the a major economic engine. It’s essential to the health of the biosphere. Agricultural products are the most popular human goods. They must not only adjust to climate change, they must also accommodate growing demand for more food of higher quality. IoT and ML technology is rapidly developing, all these have made real-time precision agriculture solution possible. This paper presents a systematic framework for agricultural crops prediction, sensors analysis, and soil parameters detection by utilizing IoT so embedded sensors and machine learning classifiers Smart field map-based Internet of Things sensors (IoT) collect sexually active real-time data on significant soil parameters like soil moisture, pH, temperature, and nutritional levels. This data is transferred to the cloud for analysis and shaping. Crop prediction is a field of research which is gaining popularity using Internet of Things (IoT). Data collected for learning is the data collected by the system through IoT sensing devices that are a classifier or predictor. This study addresses two essential points. This study presents the design, simulation and evaluation of an IoT-based soil parameter monitoring sensor. In the second run crop is predicted on not the basis of Koggle crop recommendation dataset soil factors using SVM based classifiers. With the help of Node MCU we can create IOT soil moisture sensor. Next, the relationship between the moisture and pH value is used to enrich the database attributes with moisture characteristics. Machine Learning (ML) classifiers are applied to this new hybrid data set consisting of extended information about N, P, K, temperature, humidity, pH, rainfall and estimated crop wetness to classify the appropriate crop. The 100 samples of each of the 18 crops made the database. The second model is the regression model which is used in the design procedure of the soil parameters. Just the hybrid approach alone gives 98.6% accuracy in cross validation tests.