Science Buddies: "Ask an Expert"

I have already completed my first project on KNN and its application revolving Classifying cancer cells. However, this year I want to take it to the next level. To do this I want to not just classify the cancer cells as benign or malignant but instead classify them and cancerous or not. I have tried to accomplish this but in my research I have hit a dead end. So, I need a data set with cancerous and non cancerous cells, similar to the WDBC data set. This new data set I will use in place of the WDBC in the Google co lab to achieve a KNN that can classify as cancerous and not. Also, will the new data set be applicable to the Google co lab provided in the project recourses. Ie. creating PCA charts of new data and overall just working the same way.

[Administrator note: Your other copies of this post have been removed. It's best to have your question in a new post. Breast Cancer KNN Project: https://www.sciencebuddies.org/science- ... ast-cancer ]

Hi - It is great to hear you already did the initial KNN project. Does that mean you did this project? https://www.sciencebuddies.org/science- ... ast-cancer

It's a bit unclear what you are looking to do next. You write: "To do this I want to not just classify the cancer cells as benign or malignant but instead classify them and cancerous or not."

Malignant cells are cancerous.

The "Can AI Diagnose Breast Cancer" project is about using KNN to determine if cells in the data set are malignant (cancerous) or not (benign).

We don't have another data set for you to use, but if you decide to work with data not related to breast cancer, you might look at https://www.kaggle.com/ or https://archive.ics.uci.edu/. Note that it will be easiest to work with CSV files, if you can get the data in that format.

Amy
Science Buddies

Sorry if I worded my question unclearly. My intention with the new science project I hope to complete will be similar to the science buddies version. The science buddies KNN project revolves around classifying cells as benign or malignant. I want to classify cells as cancerous or not. Unfortunately, I have not found a data set that works and my question was asking if anyone knew of one. The data set must be compatible with the Google co lab that science buddies provided.

I think the source of confusion remains that "malignant" means cancerous.

It sounds like you are looking to do the same thing just with different words.

Classifying cells as benign or malignant is the same as classifying the cells as cancerous or not.

Beyond the sources previously listed where you could pull a different data set, we don't have another data set to recommend for plugging into the project.

Amy
Science Buddies

I see the mistake in my wording. Im sorry, I meant if cancer is present rather benign or malignant.