Orange: Visalization of Data Subsets
Sumber: https://orange.biolab.si/workflows/
Beberapa widget visualisasi, seperti Scatter Plot dan beberapa widget proyeksi data, dapat mengekspos instance data dalam subset data. Dalam workflow ini, Scatter Plot memvisualisasikan data dari file data input, tetapi juga menandai titik data yang telah dipilih dalam Tabel Data (baris yang dipilih).
The workflow in Orange Data Mining shown in the image follows a basic data selection and visualization approach using a Data Table and Scatter Plot. Here’s the step-by-step breakdown:
1. File (Data Loading)
- The File widget loads a dataset into Orange.
- This dataset contains multiple instances (rows) with different attributes (columns).
2. Data Table (Selecting Data Instances)
- The Data Table widget is used to explore the dataset.
- Users can manually select specific data instances or a subset of rows using the Shift key.
- This selection helps in filtering data for further analysis.
3. Data Subset Connection (Verifying Selection)
- The Selected Data → Data Subset connection ensures that only the chosen subset from the Data Table is passed to the next widget.
- Users can double-click on this connection to verify that the data subset is correctly transferred.
4. Scatter Plot (Visualizing Selected Data)
- The Scatter Plot widget is used to visualize the selected data subset.
- Users can observe the relationships between different variables in a 2D scatter plot.
- This helps in analyzing patterns, trends, and possible correlations between features.
Summary
This Orange Data Mining workflow loads a dataset, allows for manual selection of specific data instances in a Data Table, and visualizes the selected subset using a Scatter Plot. It is useful for exploratory data analysis, interactive selection, and visual pattern recognition.
Source