Ứng dụng kỹ thuật phân loại Logistic Regression hỗ trợ chẩn đoán bệnh tiểu đường thông qua dữ liệu

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Thị Kim Dung Đỗ

Ngôn ngữ: Vie

Ký hiệu phân loại: 610 Medicine and health

Thông tin xuất bản: Tạp chí Công thương 2023

Mô tả vật lý: 348-353

Bộ sưu tập: Metadata

ID: 397050

Thêm vào giỏ Liên kết toàn văn

Tóm tắt
Chủ đề

Hiện nay, nhu cầu khám chữa bệnh tiểu đường ở các cơ sở y tế đang ngày càng tăng cao, đặc biệt bệnh nhân khám về bệnh này đa phần tập trung chủ yếu vào buổi sáng. Một ứng dụng kỹ thuật phân loại trong học máy hỗ trợ chẩn đoán bệnh tiểu đường có độ chính xác cao sẽ giúp người bệnh biết được kết quả sớm và tiết kiệm thời gian, nhân lực cho ngành Y tế. Bài báo này trình bày về đặc trưng của kỹ thuật phân loại Logistic Regression (LR) trong học máy và sử dụng bộ dữ liệu Diabetes để thử nghiệm đánh giá chẩn đoán bệnh tiểu đường qua một số độ đo hiệu suất như độ phân loại chính xác, độ chính xác, độ bao phủ, độ đo Fl score và AUC. Kết quả thử nghiệm cho độ chính xác của LR là 79%. Dựa vào kết quả này, việc ứng dụng kỹ thuật phân loại LR hỗ trợ chẩn đoán bệnh tiểu đường thông qua tập dữ liệu sẽ mang lại hiệu quả., Tóm tắt tiếng anh, Currently, the demand for diabetes examination and treatment in healthcare facilities is increasing. Especially, patients mainly come to healthcare facilities on the morning. The application of the classification techniques in machine learning to support the diabetes diagnosis with a high accuracy will help patients know the results early and also help healthcare facilities save time and human resources. This study presents the characteristics of the Logistic Regression, a classification technique, of machine learning. In this study, a Diabetes dataset is used to test the diagnosis of diabetes by using the Logistic Regression. The performance of this classification technique is assessed through some measures, such as: the classification accuracy, the precision, the coverage, the Fl score and the AUC. The test results show that the Logistic Regression's accuracy is 79 percent. This result confirms that the application of the Logistic Regression will effectively support the diabetes diagnosis.

Tạo bộ sưu tập với mã QR