Oracle Data Science Data Visualization

Oracle Data Science Data visualization (DV) is a good way to do data exploration and data analysis in data science work (exploratory data analysis (EDA)). It enables analysts to efficiently derive insights from data. It enables data science to identify and understand patterns and relationships quickly and easily. And no special training is required to interpret what is shown in the graph.

Automatic Visualization

ADS provides automated visualizations that help data scientists generate insights quickly and efficiently. Automatically detects the type of your data columns and generates summary statistics, distribution charts, co-correlation maps, and anomalous data profiles

Automatic Visualization: Methods

corr( ) : To visualize the correlation, call the show_corr() method. If the correlation matrix has not been cached, this call triggers the corr() function, which computes the correlation matrix.
- Continuous-Continuous : It uses the Pearson correlation coefficient method. It is a measure of linear correlation between two sets of data. The correlation range is from – 1 to 1.
- Continuous-Categorical : It uses correlation ratio. It is a measure of the curvilinear relationship between the statistical dispersion within individual categories. The correlations range from 0 to 1.
- Categorical-Categorical : It uses Cramer’s V method. It is a measure of association between two nominal variables, which gives a value between 0 and 1.
show_in_notebook( ) : It creates a comprehensive preview of all the basic information about a data set, including:
- Type of data set (regression, classification, or multiclass classification)
- Number of columns and rows, and feature types of each column
- Correlation map and data set header
plot ( ) : The plot() method is an automatic plotting tool to explore the relationship between two columns. pass in a variable for the x axis and, optionally, a variable for y, and then call show_in_notebook()to plot : ds.plot(“col”).show_in_notebook(figsize=(4,4))
feature_plot( ) : The feature type system in ADS provides plots for all ADS-supported feature types. It’s easy to create feature plots for your custom feature types. Calling .feature_plot() on a Pandas series produces a univariate plot, calling it on the entire dataframe returns a dataframe where each row represents a feature.
- series.feature_plot() creates a single plot for that feature.
- df.feature_plot()creates a collection of plots for all the features in the dataframe.

Customized Visualization

ADS gives you several options for plotting the columns of the dataset. The visualization API is flexible enough that you can customize the chart or choose your own plotting library.
You can use the ADS Call() method to choose your own drawing program. The call() method allows users to draw using their favorite drawing library/package in a more flexible way.

Seaborn, Matplotlib, and Geographic Information System (GIS) external packages can be used to plot custom graphs.

Seaborn : The function creates a grid of axes such that each variable in data is shared in the y-axis across a single row and in the x-axis across a single column