Insights into Cottonseed Cultivar Identification Using Raman Spectroscopy and Explainable Machine Learning
importance 4/10
ai assisted
Spectroscopy and Chemometric Analyses seed science stage: research
Problem
Traditional cottonseed cultivar identification is slow and difficult, risking inauthentic seed reaching farmers.
Why it matters
Authentic seed identity directly protects farmers' livelihoods, as stated in the abstract.
What the AI did
Machine learning algorithms (SVM, XGBoost, MLP) combined with Raman spectral fingerprints, PCA and Harris Hawk feature optimization to classify cottonseed cultivars, explained via SHAP.
Results
not stated in source
Impact
Evidence supports a working Raman-plus-ML identification framework, but no quantitative accuracy is given.
Record
| Venue | Agriculture |
|---|---|
| Published | 2023 |
| Work type | article |
| Authors | Jianan Chi, Xiangxin Bu, Xiao Li Zhang, Lijun Wang, Nannan Zhang |
| Citations | 20 (OpenAlex, at import time) |
| References | 49 |
| Model or method | Support Vector Machine, XGBoost, Multi-Layer Perceptron |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/agriculture13040768 |
| Open access | gold |
| Keywords | Raman spectroscopy, machine learning, cottonseed, cultivar identification, SHAP, XGBoost |
Importance rating
4
real but narrower advance
Abstract states authentic cottonseed identity is crucial for preserving farmers' livelihoods and that traditional identification is slow and difficult; a real agricultural problem, but narrow and unquantified.