Agriculture is essential to the Indian economy. Population growth faces the most serious threat to food stability. Population growth raises demand, facing farmers to produce more to keep up with the demand. Crop yield prediction technology can assist ranchers in improving efficiency and productivity. Correct manure rates are required for the cultivation of oilseed crop yield. When nutrients are scarce or over-fertilized, yields are significantly reduced and the environmental burden is enhanced. To resolve these concerns, our proposed work employs machine learning techniques in the prediction of the yield of oilseed crops using organic manure as well as the amount and type of agricultural manure to be used for a specific crop in different districts of Tamil Nadu. The training set consists of actual yield data from 1961 to 2007 and the validation set consists of data from 2008 to 2019. The proposed algorithm‟s results are compared to those of other machine learning algorithms namely bagging, random forest, linear regression and naive bayes with accuracy rates of 98.5%, 96.5%, 94.5%and 92.5% respectively. According to the study, bagging (Bootstrap Aggregation) outperforms other algorithms for crop yield prediction, while the boosting algorithms perform better for recommendation systems for determining which crop to plant, which type of organic manure to use and how much manure to use in a specific area and time.