Orange: Correlations
Sumber: https://docs.biolab.si//3/visual-programming/widgets/data/correlations.html
Hitung semua korelasi atribut berpasangan.
Input
Data: input dataset
Output
Data: input dataset Features: selected pair of features Correlations: data table with correlation scores
Korelasi menghitung skor korelasi Pearson atau Spearman untuk semua pasangan fitur dalam dataset. Metode-metode ini hanya dapat mendeteksi hubungan monoton.
- Correlation measure:
- Pairwise Pearson correlation.
- Pairwise Spearman correlation.
- Filter for finding attribute pairs.
- A list of attribute pairs with correlation coefficient. Press Finished to stop computation for large datasets.
- Access widget help and produce report.
Contoh
Korelasi dapat dihitung hanya untuk feature numerik (continuous), jadi kita akan menggunakan housing sebagai contoh dataset. Load di File widget dan hubungkan ke Correlations. Pasangan fitur yang berkorelasi positif akan berada di bagian atas daftar dan berkorelasi negatif akan berada di bagian bawah.
Go to the most negatively correlated pair, DIS-NOX. Now connect Scatter Plot to Correlations and set two outputs, Data to Data and Features to Features. Observe how the feature pair is immediately set in the scatter plot. Looks like the two features are indeed negatively correlated.
