Orange is all about data visualizations that help to uncover hidden data patterns, provide intuition behind data analysis procedures or support communication between data scientists and domain experts.
Visualization widgets include scatter plot, box plot and histogram, and model-specific visualizations like dendrogram, silhouette plot, and tree visualizations, just to mention a few.
Any such interaction will instruct visualization to send out a data subset that corresponds to the selected part of visualization.
Scatter plot is great for visualizing correlations between pair of attributes, box plot for displaying basic statistics, heat map to provide an overview across entire data set, and projection plots like MDS for plotting the multinomial data in two dimensions.
Interactive visualizations enable exploratory data analysis.
One can select interesting data subsets directly from plots, graphs and data tables and mine them in them downstream widgets.
Say, when data has many features, which feature pair should we visualize in a scatter plot to provide most information? Intelligent visualization comes to the rescue! In Orange’s scatter plot, this is called Score Plots.
This article was summarized automatically with AI / Article-Σ ™/ BuildR BOT™.