Hi,
I am Nguyen. I am an advisory of a big firm
I understand your needs as below ( please correct me if anything wrong):
- You have a dataset which contains attibutes and result ( breast cancer : yes or no) of patients.
- You need a thesis and source code to classify which one will be breast cancer or not.
From my point of view, two things should be done:
- From historical data, we use some methods to classify which one with specific attributes is breast cancer. It could be done by clustering, decision tree, ... But that is only the 1st part of the story. It only tells you the truth of past. It does not tell you a patient come with certain attribute is likely breast cancer or not in a period time observation (Ex: in the next 6months or 12 months) . It only looks backward.
- So If you have enough data, a model should be built to estimate probability of breast cancer of a patient with specific attributes now in next period time observation. It's the second part of the story. It
tells the story of future. It looks forward.
From my experience, your requirements can be completed in five days if data is available and in spreadsheet form. If we have a deal and after I saw your data, I can tell you exactly which thesis should be used and also source code transfer to you when I complete the project.
I have done some projects with similar requirements so I am confident to finish it on time with a condition is data available from you.
Thank you and regards,
Luan Nguyen.