1
$\begingroup$

I'm working on an SVM model as my college project. And the goal is to identify whether a tumor is benign or malignant. I'm implementing the model in Python. I found the data set from Gene Expression Omnibus.

Link to data set. The data set is related to Thyroid cancer. As you scroll down you can see a table. My problem is how can I feed that data to my model.

I don't have much idea on dealing with the data set. Could you tell me more about the data set from the mentioned website? I didn't understand it properly.

I'd be happy if you help me out with this. I'm open to all kinds of suggestions and ideas. If you know about any another data set repos, let me know.

Have a great day.

$\endgroup$
  • 2
    $\begingroup$ I think this is a good example of how lost one can be without domain knowledge :-) But anyway, It seems the sequences can be ignored and simply treat them as categorical values and onehot encode them if they arent to discriminate/unique. But there might be important information in thier structure you miss out on. $\endgroup$ – CodeMonkey Jan 30 '19 at 8:17
  • $\begingroup$ Hi, there - Mr. @CodeMonkey. Thank you for taking your time and I appreciate it. I was looking for something more insightful. Like a description of the fields, values in the data set. And what the data set is about in an insightful way. Yes, my question is a good example of how one can be lost without domain knowledge. Thanks for pointing that out. Have a great day. $\endgroup$ – yesiamabdul Jan 30 '19 at 12:36
1
$\begingroup$

you can feed data as SVM(train_X,train_Y) before feeding the data to model,do required proprocessing of data like having dummies or one-hot encoding for categorical variables and scaling numerical variables when mean of numerical column values are 10fold or more apart

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.