Applicability of Machine Learning Algorithms for Intelligent Farming
| Autor: | Nikhil Sharma, Bharat Bhushan, Ila Kaushik, Bharti Verma |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Boosting (machine learning)
business.industry Computer science Decision tree learning Feature extraction Decision tree Machine learning computer.software_genre Support vector machine Data analysis Production (economics) Artificial intelligence business computer Algorithm Statistical hypothesis testing |
| Zdroj: | Studies in Big Data ISBN: 9783030756567 |
| DOI: | 10.1007/978-3-030-75657-4_6 |
| Popis: | Agriculture contributes enormously to the growth and economy of a country due to which it becomes important to upgrade the agricultural facilities for farmers that simulate them for cultivating good quality crops with high production rates. This paper sets sights on classifying different types of crops grown, and predicts which crop is best suited for particular location for boosting the production factor. Further, this ML model will be integrated with Internet of Things (IoT) to build an intelligent irrigation system that itself decides whether the crop-land needs to be irrigated or not. This system uses decision tree algorithm, Arduino, sensors, and bolt IoT kit. By means of feature extraction and data analysis techniques, we were able to select highly meaningful and best contributing variables from gathered data that were affecting the prediction values. Also, we discovered and unleashed the working statistics behind certain powerful ML algorithms. Strong statistics like hypothesis testing, chi-square testing and Euclidean distance are thoroughly discussed. Different classification models like K-NN, decision tree, SVM (Support Vector Machine) and logistic regression were implemented and compared in order to reach the best suited model for forecasting the crop class label. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|