Fertilizer Recommendation System using SGD on Mahout and Hadoop Platform
Raghu Garg1, Himanshu Aggarwal2
1Raghu Garg, Computer Engineering/ Punjabi University, Patiala /India
2Dr. Himanshu Aggarwal, Computer Engineering/ Punjabi University, Patiala /India.
Manuscript received on 30 June 2019 | Revised Manuscript received on 05 July 2019 | Manuscript published on 30 July 2019 | PP: 620-624 | Volume-8 Issue-9, July 2019 | Retrieval Number: I7571078919/19©BEIESP | DOI: 10.35940/ijitee.I7571.078919
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In the current world scenario, the existence of human is impossible without the necessary proliferation of plants. Health of plant depends on water and soil nutrition that help plants to produce energy. Apply appropriate recommended fertilizer quantity is necessary for a healthy plant. However, due to overexposure, soil sometimes gets degraded, so fertilizer is an important element to retain the soil quality. Now a days decision support system plays a vital role in the recommendation. These recommendation systems are based on historical data. In this respect, soil analysis is an appropriate approach to determine the soil quality. Soil analysis generates a report of unstructured and unperceivable data by testing soil in laboratories that make it agriculture big data. This type of systems generally has been implemented in the banking and health care sector for fraud detection and patient recommendations, respectively. In this paper, we have been proposed fertilizer recommendation system based on present nutrition quantity in the soil. In this system, the useful data is extracted from soil analysis reports and save into two files: 1) first file save soil nutrition composition and solution number that act as the label in 2) second file save solution number and recommended fertilizer quantity. Soil composition encoded into vector use by classification system to trained system. In this research work, SGD big data analysis machine learning techniques are applied to identify the fertilizer recommendation classes based on present soil nutrition composition. Here, SGD classification system is used to train the system. Our proposed system obtained 64.08% total average accuracy. The proposed model can also be used by agriculture experts to recommend fertilizer quantity according to crop type and present nutrition composition.
Keywords: Agriculture Industry, Big Data Analytics, Fertilizer Recommendations, Hadoop.
Scope of the Article: Big Data Analytic