Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM63166
Volume:
10 |
5 |
May 2026
Pages:
1-3
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
COMPUTER SCIENCE AND ENGINEERING (AI & ML)
Rajeev Institute of Technology, Hassan
COMPUTER SCIENCE AND ENGINEERING (AI & ML)
Rajeev Institute of Technology, Hassan
COMPUTER SCIENCE AND ENGINEERING (AI & ML)
Rajeev Institute of Technology, Hassan
COMPUTER SCIENCE AND ENGINEERING (AI & ML)
Rajeev Institute of Technology, Hassan
COMPUTER SCIENCE AND TECHNOLOGY (ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING)
Rajeev Institute of Technology, Hassan, 573201, Karnataka
Agriculture is the cornerstone of food security and economic stability, particularly in developing nations such as India. Selecting the right crop is a complex decision influenced by soil composition, climate, and environmental factors. This paper presents a dual-mode intelligent crop recommendation system combining classical machine learning with Convolutional Neural Network (CNN)-based deep learning. For structured inputs — soil nutrient levels (N, P, K), temperature, humidity, rainfall, and pH — three classifiers were trained: Decision Tree, Support Vector Machine (one-vs-all), and Random Forest. Random Forest achieved the highest classification accuracy of 99.55% on the test split. For unstructured inputs, a CNN model classifies soil types from images and recommends the top-three suitable crops with confidence scores. Experimental results confirm that the hybrid approach outperforms single-modality baselines.
KEYWORDS
This paper presented a dual-mode intelligent crop recommendation system combining machine learning and CNN- based deep learning. Among three ML classifiers evaluated, Random Forest achieved the highest test accuracy of 99.55%. The CNN image classification branch achieved ~88% validation accuracy and successfully generated top-3 crop recommendations with confidence scores. The hybrid approach delivers more flexible, informative, and robust outputs than any single-modality approach, and can meaningfully support farmers in evidence-based crop selection, ultimately contributing to improved agricultural productivity and food security.
[1] R. Sujatha and P. Isakki, "Crop Yield Forecasting Using Machine Learning Algorithms," Int. J. Eng. Research & Tech., vol. 9, no. 3, pp. 1– 5, 2020. [2] S. Ramesh and R. Vardhan, "Analysis of Crop Prediction Using Machine Learning Techniques," Int. J. Scientific & Tech. Research, vol. 8, no. 10, pp. 1–6, 2019. [3] L. Breiman, "Random Forests," Machine Learning, vol. 45, no. 1, pp. 5– 32, 2001. [4] K. Patel, M. Shah, and P. Jain, "Soil Classification Using CNNs for Smart Farming," Procedia Computer Science, vol. 167, pp. 198–205, 2020. [5] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, 2016. [6] F. Pedregosa et al., "Scikit-learn: Machine Learning in Python," JMLR, vol. 12, pp. 2825–2830, 2011. [7] A. Kamilaris and F. X. Prenafeta-Boldú, "Deep Learning in Agriculture: A Survey," Computers and Electronics in Agriculture, vol. 147, pp. 70– 90, 2018. [8] S. P. Mohanty, D. P. Hughes, and M. Salathé, "Using Deep Learning for Image-Based Plant Disease Detection," Frontiers in Plant Science, vol. 7, 2016. [9] P. Sharma, A. Kumar, and R. Singh, "Crop Prediction Using Machine Learning," in Proc. ICCCS, 2020, pp. 45–50. [10] D. Khaki and L. Wang, "Crop Yield Prediction Using Deep Neural Networks," Frontiers in Plant Science, vol. 10, 2019.
Publication Date
5/14/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
2
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...