Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM69049
Volume:
10 |
11 |
Nov 2026
Pages:
1-5
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 Engneering(Data Science)
SJB Institute Of Technology

Computer Science and Engneering(Data Science)
SJB Institute Of Technology

Computer Science and Engneering(Data Science)
SJB Institute Of Technology

Computer Science and Engneering(Data Science)
SJB Institute Of Technology
Crop selection and fertiliser decisions in smallholder farming still rely on experience and occasional laboratory soil tests, so field conditions that change within days go unexamined. This paper presents the implementation of a Smart Precision Agriculture System that connects soil and weather sensing, machine-learning crop recommendation and rule-based advisory in one Flask web application backed by MySQL. An IoT sensor simulator streams eight parameters (nitrogen, phosphorus, potassium, temperature, humidity, soil moisture, pH and rainfall) to a relational store every 30 seconds. A soft-voting ensemble of Random Forest, Support Vector Machine, Gaussian Naive Bayes and Logistic Regression classifies the seven agronomic features into 22 crops and returns the three most probable crops with predicted class probabilities. A deterministic rule engine produces nutrient-deficiency advice and raises alerts for low soil moisture, acidic or alkaline pH and high temperature. On the public Crop Recommendation dataset (2,200 samples, 22 balanced classes) the ensemble reaches 99.32% accuracy and 99.37% macro-precision on a stratified 20% hold-out set and 99.41% under five-fold cross-validation; top-3 accuracy is 100%. These results demonstrate the functionality of the implemented software pipeline on the curated benchmark dataset but do not establish real-world agricultural effectiveness or field-level recommendation accuracy. It does not establish field accuracy, because the dataset is a curated benchmark and the sensor stream in this prototype is simulated. Both limitations are stated and addressed in the future-work plan.
KEYWORDS
This paper presented a Smart Precision Agriculture System that unifies a soil and weather data stream, a four-model soft-voting crop recommender, rule-based fertiliser advice and condition alerts behind a Flask and MySQL web application. On the Crop Recommendation dataset the ensemble reaches 99.32% hold-out accuracy, 99.41% cross-validated accuracy and 100% top-3 accuracy, and its top-three output with confidence scores lets a farmer weigh alternatives. The ensemble does not outperform Random Forest or Naive Bayes on this benchmark, and the benchmark itself is too separable to predict field performance. The contribution is a working, tested integration whose components can each be replaced with real data or better models. Future work will (1) replace the simulator with physical NPK, moisture, pH and DHT-class sensors reporting over MQTT or HTTP; (2) retrain on regionally collected and agronomist-validated data and report per-class precision and recall; (3) connect live weather forecasts to supply rainfall and temperature trends instead of a single reading; (4) add SHAP-based explanations [13] so each recommendation shows which nutrients and climate variables drove it; (5) connect the crop-specific CSV fertiliser rules and add dose guidance validated against state agricultural guidelines; (6) add irrigation scheduling from soil-moisture trends; and (7) add authentication, secret management and a production server before field deployment.
[1] N. Zhang, M. Wang, and N. Wang, “Precision agriculture—a worldwide overview,” Computers and Electronics in Agriculture, vol. 36, nos. 2–3, pp. 113–132, 2002, doi: 10.1016/S0168-1699(02)00096-0. [2] A. McBratney, B. Whelan, T. Ancev, and J. Bouma, “Future directions of precision agriculture,” Precision Agriculture, vol. 6, no. 1, pp. 7–23, 2005, doi: 10.1007/s11119-005-0681-8. [3] O. Elijah, T. A. Rahman, I. Orikumhi, C. Y. Leow, and M. H. D. N. Hindia, “An overview of Internet of Things (IoT) and data analytics in agriculture: Benefits and challenges,” IEEE Internet of Things Journal, vol. 5, no. 5, pp. 3758–3773, Oct. 2018, doi: 10.1109/JIOT.2018.2844296. [4] M. Ayaz, M. Ammad-Uddin, Z. Sharif, A. Mansour, and E. H. M. Aggoune, “Internet-of-Things (IoT)-based smart agriculture: Toward making the fields talk,” IEEE Access, vol. 7, pp. 129551–129583, 2019, doi: 10.1109/ACCESS.2019.2932609. [5] A. Tzounis, N. Katsoulas, T. Bartzanas, and C. Kittas, “Internet of Things in agriculture, recent advances and future challenges,” Biosystems Engineering, vol. 164, pp. 31–48, Dec. 2017, doi: 10.1016/j.biosystemseng.2017.09.007. [6] K. G. Liakos, P. Busato, D. Moshou, S. Pearson, and D. Bochtis, “Machine learning in agriculture: A review,” Sensors, vol. 18, no. 8, Art. no. 2674, Aug. 2018, doi: 10.3390/s18082674. [7] A. Kamilaris and F. X. Prenafeta-Boldú, “Deep learning in agriculture: A survey,” Computers and Electronics in Agriculture, vol. 147, pp. 70–90, Apr. 2018, doi: 10.1016/j.compag.2018.02.016. [8] A. Ingle, “Crop Recommendation Dataset,” Kaggle. Accessed: Nov. 1, 2026. [Online]. Available: https://www.kaggle.com/datasets/atharvaingle/crop-recommendation-dataset [9] T. G. Dietterich, “Ensemble methods in machine learning,” in Multiple Classifier Systems, MCS 2000, Lecture Notes in Computer Science, vol. 1857, pp. 1–15. Berlin, Germany: Springer, 2000, doi: 10.1007/3-540-45014-9_1. [10] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, Oct. 2001, doi: 10.1023/A:1010933404324.
Publication Date
11/17/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
1
Total views since publication
PDF Downloads
1
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...