Crop Recommendation System Using Machine Learning

Agriculture is considered as primary profession and one of the significant way of life in Pakistan. It plays a major part in the economy of Pakistan. A huge amount of Pakistan’s people are dependent on agriculture. Agricultural land can be utilized efficiently and effectively to ensure food security in Pakistan. Hence it is important to enhance yield of crop. Cultivation of the crops throughout the world rely on the soil and environmental conditions. It is crucial to select appropriate crop for cultivation by considering soil and environmental factors to improve production of crop. However, farmers don’t select best and accurate crop for cultivation due to unawareness about environmental and soil properties. In this way production of crop suffers. Now a days emerging technologies have a major and special role in various sectors to resolve the problems, make things easy and to give high accurate best results. Similarly, these advance technologies such as machine learning and deep learning can be utilized properly in the field of agriculture to overcome challenges. Implementing latest technological techniques in the agricultural sector is beneficial in this current situation. To address the challenges of enhancing crop yield in Pakistan. A smart crop recommendation system was developed by using machine learning technology, in this study to suggest favorable crop for cultivation. This proposed system can be more beneficial for farmers to upgrade production of crop. In this proposed system, various machine learning algorithms were utilized for training the model including Decision Tree (DT), Support Vector Machine (SVM) and Logistic Regression (LR). Decision Tree algorithms given best performance with accuracy of 99% than Support Vector Machine (SVM) and Logistic Regression (LR). Hence Decision Tree algorithm was included for this proposed system. This designed system suggests most profitable crop to the farmers by taking environmental and soil properties such nitrogen (N), phosphorus (P), potassium (K), pH level, temperature, humidity and rainfall.