Skripsi
ANALISIS KOMPARATIF KETEPATAN HASIL DETEKSI KANKER PROSTAT MENGGUNAKAN METODE RANDOM FOREST DAN K-NEAREST NEIGHBOR (KNN)
Prostate cancer is a malignant disease that typically affects men and occurs in the prostate gland located beneath the bladder. Prostate cancer ranks as the second most common cancer in the United States and was the fifth in Indonesia with 13,563 cases in 2020. While the exact cause remains unknown, early detection of prostate cancer can be achieved through the use of machine learning and data mining techniques. Data were sourced from Kaggle, incorporating ten features. The Random Forest and K-Nearest Neighbor (KNN) methods were employed for classification, with PCA used to select eight components corresponding to the number of features in the training data. The research findings reveal that KNN outperformed Random Forest in classification performance. In KNN, the optimal parameter K was identified as 19, achieving an accuracy, precision, recall, and F1 score of 100%. In contrast, Random Forest attained an accuracy of 75%, with precision and recall at 85% and 75%, and an F1 score of 75%. These results indicate that KNN can classify data with higher accuracy. The findings demonstrate the efficacy of both methods in classifying prostate cancer patients.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2407000501 | T137618 | T1376182024 | Central Library (Referens) | Available but not for loan - Not for Loan |
No other version available