Site-Specific Crop Management

Machine learning (ML) is an emerging technology in which both researchers and academic communities have been able to obtain a distinct range of numerous applications such as image recognition, healthcare, fraud detection, and so on. In addition to this, ML algorithms have been identified for prediction, classification, and other purposes. On the contrary, agriculture plays an important role in a country's economy. Further crop yield is mainly affected by diverse factors like types of soil, temperature, moisture, and rainfall. Moreover, the crops are not evenly distributed in an agricultural field restricting the possibility of complete utilization of the field. Even modern technology like artificial intelligence(AI), ML, and the Internet of Things(IOT) help to overcome problems related to agricultural aspects. Therefore ML is reliable, suitable and well developed, which are not utilized properly by farmers. The proposed research work is to predict the crop yield of a given area and map the yield distribution using an ML algorithm through the Random Forest (RF) mechanism. RF approach is a widely used supervised learning technique that may be applied to regression and classification problems. Hence crop yield prediction is carried out by analyzing data on various factors affecting agriculture like temperature, soil type, moisture, etc. Subsequently, the main motivation of this research is to build an interactive user-friendly interface for the farmers which provides the yield map distinguishing the yield rate in a given field along with suggestions for the better utilization of the low-yielding areas.