Data Audit node
The Data Audit node provides a comprehensive first look at the data you bring into Watson Studio, presented in an easy-to-read matrix that can be sorted.
When you run a Data Audit node, output is generated that includes:
- Summary statistics, histograms, and distribution graphs that may be useful in gaining a preliminary understanding of the data.
- Information about outliers, extremes, and missing values.
Using the Data Audit node
The Data Audit node can be attached directly to an Import node or downstream from an instantiated Type node.
Screening or sampling the data. Because an initial audit is particularly effective when dealing with big data, you might use a Sample node to reduce processing time during the initial exploration by selecting only a subset of records. The Data Audit node can also be used in combination with nodes such as Feature Selection and Anomaly Detection in the exploratory stages of analysis.