Leveraging Machine Learning: Advanced Algorithms for Soil Data Analysis and Feature Extraction in Arid and Semi-arid Regions with Expert Systems
- Title
- Leveraging Machine Learning: Advanced Algorithms for Soil Data Analysis and Feature Extraction in Arid and Semi-arid Regions with Expert Systems
- Creator
- Gulledmath S.; Hemanth K.S.
- Description
- India is culturally diverse nation at large. There are two words of symphony one is tradition and second one is inherited agriculture. India has long historical advantage having conventional agricultural practices and the scope for it to dive into day to day life as agriculturist. Happiness shrinks as people grow into modern world current trend of agriculture faces a monument challenge and needs immediate address to survive. Now withstanding with this phrase of human life on earth its necessary to give more importance to soil rather than the existence. Soil health is more paramount in this equation, as it directly influences crop growth and yield. Traditionally, analysing a few key soil properties has been the cornerstone of soil treatment practices. However, this approach often overlooks the complex interplay between various soil characteristics. To overcome the above hurdle present research incorporates the method of multivariate data analysis with selective advanced algorithms in machine learning to find suitability to predict best fit algorithm in real time data sets in arid and semi-arid zones of kolar district in Karnataka. The purpose is to draw the attention of stake holders to leveraging the new technology to deploying them into effective assessment in building expert system to incorporate in regular use on handy devices. This penetrates the results by two extremely good classifications algorithms Decision Tree and Gradient Boosting emerged as winner with 99% accuracy. In contrast, Passive Aggressive and Linear SVC produced below average of 36% accuracy. The ensemble algorithms of SMOTE on Random Forest and Stochastic Decent Gradient produced the acceptable accuracy of 83%. This input helped dynamically to build ready to use expert systems for farmers. The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.
- Source
- SN Computer Science, Vol-5, No. 7
- Date
- 2024-01-01
- Publisher
- Springer
- Subject
- Classification algorithms and expert systems; Machine learning; Precision agriculture; Soil characteristics
- Coverage
- Gulledmath S., School of CSA, REVA University, Karnataka, Bengaluru, India; Hemanth K.S., Department of Computer Science, Christ University, Yashwanthpur Campus, Karnataka, Bengaluru, India
- Rights
- Restricted Access
- Relation
- ISSN: 2662995X
- Format
- Online
- Language
- English
- Type
- Article
Collection
Citation
Gulledmath S.; Hemanth K.S., “Leveraging Machine Learning: Advanced Algorithms for Soil Data Analysis and Feature Extraction in Arid and Semi-arid Regions with Expert Systems,” CHRIST (Deemed To Be University) Institutional Repository, accessed February 24, 2025, https://archives.christuniversity.in/items/show/12817.