I wanna ask very important question about the random population generation gin splitting the dataset in machine learning classification models;
For example to more explain, i used seed =1 a,d i got accuracy of 0.7 and seed= 5 and i got accuracy of 0.8 and seed= 2000 and i got accuracy of 0.89 using Adaboost.
I found research paper using the same datase i used and accuracy achieved is 0.94 using xgboost model without specifiying the seed used in developing the model.
same for other research papers exists.
My question is which results i ave to pick to compare my model with other models proposed in literature Meanwhile i implemented all the models proposed in literature with the different seed i used and i found not the same results in their paper and sometimes with not all the seeds my result with adaboost is better.
I need help to compare my proposal with other works