Icon

preprocess_​5A (1)

10. Kiểm thử

Sau khi đã có data, kéo trực tiếp vào input (nhánh 2) của Predictor.

Execute Scorer là xong bước 10.

9. Model

1-8. Preprocessing

9.2. Ứng dụng mô hình đã chọn và so sánh dữ liệu

9.1. Chọn mô hình phù hợp

93.4
Scorer
X-Aggregator
Score mô hình đã PCANhánh dướ: Cross-validation
X-Partitioner
Table Partitioner
Scorer
Decision Tree Predictor
Decision Tree Predictor
Nhánh trên: Score khi chưa PCANhánh dưới: Cross-validation
X-Partitioner
Gradient Boosted Trees Predictor
Decision Tree Learner
loop chạy các mô hình với việc thêm/bớt nhiều biến để tìm ra tổ hợp biến có ảnh hưởng lớn nhất đến mô hình
Feature Selection Loop Start (1:1)
Decision Tree Learner
Scorer
Feature Selection Loop End
Decision Tree Predictor
X-Partitioner
Từ file biến quan trọng => lọc cột
Column Filter
X-Aggregator
xuất file xem các biến quan trọng
Excel Writer
Data đã qua xử lý logic ở code Google Collab: https://colab.research.google.com/drive/1wf6eCEGruD9h4rf10l7_p0RGb3leXNJK?usp=sharing#scrollTo=m6EiAVIarINs
CSV Reader
lọc trùng id
Duplicate Row Filter
chuẩn albumin
Rule Engine
xóa action
Column Filter
chuẩn hiv
Rule Engine
chuẩn glucose
Rule Engine
chuẩn malaria
Rule Engine
has_tobacco_exposure
Rule Engine
Drop miscarriages_or_abortions
Column Filter
alcohol_or_drug_use
Rule Engine
Decision Tree Predictor
has_violence_exposure
Rule Engine
flag_anemia
Rule Engine
Column Filter
flag_hypertension
Rule Engine
trimester
Rule Engine
flag_high_glucose
Rule Engine
Check class imbalance
Statistics
Decision Tree Predictor
90.3
Scorer
Decision Tree Learner
Table Partitioner
Random Forest Predictor
xóa cột missing
Missing Value Column Filter
Gradient Boosted Trees Predictor
PCA ko hiệu quả với mô hình phân loại
PCA
Random Forest Learner
Feature Selection Loop Start (1:1)
Gradient Boosted Trees Learner
Feature Selection Loop End
Scorer
Decision Tree Predictor
Decision Tree Learner
đặt tên bp
Column Renamer
lọc các cột
Column Filter
pulse
Rule Engine
Data đã hoàn thành xử lý. Dùng cho train model.https://drive.google.com/file/d/1nt1efKNfVKd-yCmsb4ir55ZhloJfP8Qj/view?usp=drive_link
CSV Writer
String to Number
Lọc các biến ít quan trọng
Column Filter
body temp
Rule Engine
lọc các cột
Column Filter
respiratory
Rule Engine
Data đã xử lý xong
Column Filter
String Manipulation (Multi Column)
Test: Data gốc đã được thêm rows mới và xử lý logic bằng codehttps://drive.google.com/uc?id=1WEgn7uObrzC7Tg18nLvHbiRgwBHU2BkH
CSV Reader
mean + most
Missing Value
String Manipulation (Multi Column)
Từ file biến quan trọng => lọc cột
Column Filter
unknown
Missing Value
xóa hằng số
Constant Value Column Filter
chuẩn urine
Rule Engine
mean
Missing Value
String Manipulation (Multi Column)
đặt tên bp
Column Renamer
String Manipulation (Multi Column)
tách bp
String Splitter (Regex)
Data đã preprocessing.(Có thể nối thẳng vào nhưng để hạn chế rối mắt nên trung gian qua csv)https://drive.google.com/file/d/1nt1efKNfVKd-yCmsb4ir55ZhloJfP8Qj/view?usp=drive_link
CSV Reader
xóa hằng số
Constant Value Column Filter
chuẩn urine
Rule Engine
chuẩn albumin
Rule Engine
lọc trùng id
Duplicate Row Filter
chuẩn hiv
Rule Engine
xóa action
Column Filter
Feature Selection Loop Start (1:1)
X-Aggregator
Scorer
chuẩn glucose
Rule Engine
Scorer
respiratory
Rule Engine
chuẩn malaria
Rule Engine
Gradient Boosted Trees Learner
Column Filter
Feature Selection Loop End
alcohol_or_drug_use
Rule Engine
pulse
Rule Engine
String to Number
99.7
Scorer
xóa cột missing
Missing Value Column Filter
Decision Tree Learner
mean + most
Missing Value
Drop miscarriages_or_abortions
Column Filter
92
Scorer
Data gốc (gốc của gốc)https://drive.google.com/file/d/107N3GYV2E8kVyI0D0cEjgrEhedNigk5w/view?usp=drive_link
CSV Reader
has_tobacco_exposure
Rule Engine
Decision Tree Learner
has_violence_exposure
Rule Engine
body temp
Rule Engine
mean
Missing Value
tách bp
String Splitter (Regex)
unknown
Missing Value

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: Dang
Created at: 2025-10-17
On NodePit since: 2025-10-23
Last update: 2025-11-14
Created with KNIME version: v5.8.0

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!