IJIRMPS | Volume 7, Issue 6, 2019 ISSN: 2349-7300 IJIRMPS1906016 Website : www.ijirmps.org Email : editor@ijirmps.org 51 Rainfall Based, Crop Production Identification Using Data Mining Swarda Joshi 1 , Apurva Kulkarni 2 , Asmita Temkar 3 , Laleeta Suryavanshi 4 , Prof. S. B. Jadhav 5 Abstract: Rainfall is depends on various factors such as temperature, humidity, cloudiness, wind speed, etc. Rainfall prediction is a major concern for many fields like meteorological department, agriculture department etc which is directly associated with the economy and sustenance of human life. In the proposed system we are going to predict the rainfall and after prediction of rainfall find out the soya bean disease occurrences if there is any. Agriculture study is fast growing, due to innovation in technologies and upcoming challenges. It has been proven to be leading role in improving the on the whole growth rate of any country. Agriculture matters a lot in the economic growth. Very wide work is going on from last so many years to improve the productivity by using data mining & evolutionary techniques. The loss is a large issue for agriculture, because it will make it more difficult to produce enough food. There is a limited analysis performed for loss of crops, due to disease or growth of insects. The correct prediction of loss in crops helps farmer to apply some suitable action to overcome the issues and maintain its expected production. As the agriculture matters a lot in the economic growth to overcome this issue we have used the data mining approach i.e classification. The objective of the system is to find the disease on crop and predict the disease according to the rainfall. So for the classification purpose algorithm is used. Data mining techniques help in agriculture for the exclusion of manual jobs and for decision making which enable to decrease production cost and improves productivity. Keywords: Crop, Classification, Naïve Bayes, KNN, RF Introduction: The agricultural study has strengthened the optimized economical income, internationally and is very huge and important area to gain more benefits. However, it can be enhanced by the use of different technological resources, tool, and procedures. Today, the term data mining is an interdisciplinary process of analyzing, processing and evaluating the real-world datasets and prediction on the basis of the findings. Data mining is the collection of exponentially growing techniques which are used to find some useful information, patterns, and knowledge from already given data. This useful information helps to advance existing research and improve productivity. The applications of data mining are uncountable; it is used almost in every aspect of life. Some major applications of data mining are Healthcare, Market Analysis, Finance, Education, Manufacture Engineering, Corporation Surveillance, and Agriculture. Lots of technology and research have been grown in the field of agriculture. To improve the rate of agriculture, the researcher has used data mining technique to solve problem-related agriculture. In this paper data mining technique is used which will predict the rainfall and crop disease. It supports the farmer while taking the right decision about agriculture. This paper focuses on the prediction of crop diseases and rainfall using datasets. For the classification and the prediction purpose the Naïve Bayes Knn, Rf algorithm is used in the proposed system. Related work: Agriculture’s input in GDP is important for many countries, especially for Asian countries. Researchers are doing very wide work from last few years to enhance the productivity by using DM & evolutionary techniques. In last few years, the question again arise “Does agriculture matters for economic growth in developing countries?” In [11] researchers have proved with output that agriculture is a backbone of economic growth for many countries. The prediction of diseases and loss in crop help farmers to improve it. There are many different classifiers used to build a model for predicting the spray decision on crops [12-17]. In [12] author also used ensemble models Random forest to improve the accuracy of weak classifiers. In [13] authors have used Decision Tree, Support Vector Machine and Neural Network classifiers. They concluded that theses classifier have good prediction accuracy as compared to Kth Neural Network and GNB. In [14] author proposed different matrices for evaluation such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and accuracy for both true & false classes in all mentioned classifiers. Author also describes that the use of ensemble model is a future research direction. In [18] author articulated that different classifiers can be used on different features in ensemble model for predicting true class. Each model for ensemble classifier trained on different set of features to achieve the best results. And high vote getting class will be allocated to test sample. In [19] authors provide an extensive survey on classifiers and explain the ways for better prediction. Authors explained that we cannot solve all the problems with same classifier. Different classifiers work in different way so different classifiers can be used in different scenarios. In following papers [20-22] Neural Network is used as classifier for predicting the diseases in fruits. Dhawal Hirani, Nitin Mishra [1] "A Survey on Rainfall Prediction Techniques". This paper mainly discusses the various machine learning techniques used for early prediction of rainfall and some cognitive approaches. Research’s mainly explained about approaches to empirical and dynamical Methods. Pinky Saikia Dutta, Hitesh Tahbilder [2] "Prediction of Rainfall Using Data Mining Technique over Assam". In this paper, they have described data mining technique in forecasting monthly Rainfall and traditional statistical technique Multiple Linear Regression.