Data Quality Metric Explanations
CluedIn calculates data quality metrics based off all data that is ingested into the platform. There are many times where you will want a verbose explanation on why a certain data quality score was given. For example, imagine if your had what you think is golden records, but CluedIn gave it very low quality scores - you would want to know exactly why.
The first option is to visualize this explanation within the user interface by clicking on the “Explain” button when you see a data quality score widget.
The second example would be to get this data from our GraphQL endpoint.
Here is an example that will fetch a detailed explanation of the “Accuracy” metric for Global and Entity Level dimensions.
metrics
{
name
dimensions
{
id
type
... on GlobalMetricDimension
{
children
{
id
}
}
... on MetricDimension
{
valueHistogram
}
}
globalLevel: dimension(id : "25118d8c-696e-591f-9eb5-083bff9488d4")
{
id
... on GlobalMetricDimension
{
children
{
id
}
}
}
entityLevel: dimension(id: "b545b625-5a9b-5614-ae57-f2c61e02bdb9")
{
id
... on MetricDimension
{
valueHistogram
}
}
globalDimension
{
id
latestValue
parentId
children
{
id
parentId
type
detailType
detail
providerDefinitionId
providerId
providerName
# valueHistogram
entityValues(
sort: VALUE,
sortDirection: ASCENDING,
cursor: "ewAiAFAAYQBnAGUAIgA6ADUALAAiAFAAYQBnAGUAUwBpAHoAZQAiADoAMgAwAH0A"
)
{
entries
{
value
dimension
{
detailType
latestValue
children
{
detailType
detail
latestValue
}
}
entity
{
name
entityType
}
}
cursor
}
children
{
id
parentId
type
detailType
detail
providerDefinitionId
providerId
providerName
valueHistogram
entityValues(sortDirection: ASCENDING)
{
entries
{
value
}
cursor
}
}
}
}
}