Bushra,
Matlab has an SVM package that you can use if you have the machine learning toolkit. The link to their documentation page is here: https://www.mathworks.com/help/stats/svmtrain.html.
The general framework here is to divide your data into a training set and a testing set. You will use the training data to train the parameters of the SVM to be later used for prediction. You can use your testing data to evaluate how well our SVM model can predict the output of unseen data.
To train the SVM model in MATLAB, you will need to define what your features are and what your output will be. For example, you can use plant height, plant color, etc. as a feature to predict what crop is growing. So, to train your SVM, create a matrix that has each example on the rows and each prediction features as the columns. This can be passed to the svmtrain function as the first parameter. The second parameter will be a vector corresponding to the group in which each example belongs. Each row of the matrix in argument one corresponds to the same row in the output vector.
The function will return a learned model that can then be used with the svmclassify function (also provided by MATLAB: https://www.mathworks.com/help/stats/svmclassify.html) to predict future events.
Hi Bushra,
I think your data is kind of time series data. So to use machine learning to forecast the future is the aproporaite way to predict.
Please check this:
https://machinelearningmastery.com/time-series-forecasting-supervised-learning/
- Badges
- Science topic