Positive and unlabeled Learning from Unbalanced cases and Sparse structures, or PLUS, represents the first one to use positive and unlabeled learning framework to specifically model the under-diagnosis issue in predicting cancer metastasis potential. PLUS is specifically tailored for studying metastasis that deals with the unbalanced instance allocation as well as unknown metastasis prevalence, which are not capable by any other methods. Its robustness grants the possibility to harness the power of big data by integrating large scale datasets from different cancer types. Insights gleaned from this research will prove useful to the diagnosis and treatment of clinical metastatic disease.
Prediction=PLUS(train_data=X,Label.obs=Label,Sample_use_time=30,l.rate=1,qq=0.1)
Details Of each parameter can be found from notations in code.
Ph.D. candidate, Indiana University School of Medicine
Assistant Professor
Department of Biostatistics, Indiana University