Objective : Classification of wheat & Mustard genotypes using support vector Machine (SVM) ?
Since SVM can classify on the basis of categorical data(Class) only like low, medium, high; but we have numerical data of wheat yield, mustard yield and Mustard oil content.
Therefore kindly suggest standard procedure to convert wheat yield, mustard yield and oil content data of 250 genotypes into categorical data(Class) data.
Hello Kailash, I would do it this way since I have done it to categorise plant type in cotton. For the traits you mentioned, yield and oil content, take the range of values (I believe you have phenotypic data of varieties/hybrids) and divide it by the number of classes you need. For eg. you can divide the range of yield by 2 to get two classes which you can call High and Low. Dividing by 3 will give you three classes as High, Medium, Low. You can call your classes by names that suit your study. This method focuses on the study values you have got and hence are relative. However, you can get some standard but loosely defined values in the literature pertaining to your crop.
https://stackabuse.com/implementing-svm-and-kernel-svm-with-pythons-scikit-learn/
Hi Sergei, do we have to learn python for devising the algorithms or can we get readymade software for at least linear classification?
- Badges
- Science topic
- Similar topics
- Engineering
- Petroleum Engineering
- Oil