hey sir, in this firstly i replace all zero with there mean or median of corresponding features i.e. removing all null values then i apply standard scaling , then I apply SMOTE method for oversampling. (i have also normalized the testing data in the same manner)But when I do testing all values predicted to be zero and getting 65 accuracy. and I tried this for many values of k . What should i do to increase the accuracy ???
please guide me.
How to increase accuracy
hey @anuj_singla8 ,
these kind of pre processing are all your own choices , any choice can make your predictions better or worse. So , think and try accordingly.
Another thing , this data is not that much unbalanced that you require SMOTE for this. You can try simply also. More things to matter for this will be model choice and feature engineering.
If it’s fixed for KNN only , then feature engineering is the only Key.
I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.
On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.