I'm a long-time Framework Manager user that's slowly easing into the world of Data Modules. And for the most part I can get things to work, but now I'm trying to figure out "how" things work. What I want to know is "how" does Cognos do the metadata processing for sources like uploaded CSV files going into a Data Module? For example, it does a good job of picking out "identifiers", "facts", and "attributes", but how does Cognos know what "usage" values to assign to those various data items that are coming out of a generic source like CSV? What metadata criteria does it use for the assignments and at what stage in the process is the criteria applied?

I checked out the IBM Cognos Analytics v11.2 Metadata Modeling Guidelines, but it doesn't drill down into the stuff I'm looking for. So does anyone out there in Cognos Land have the info that will set my mind at ease… *^_^*

Thanks in advance, for any helpful insight.

Syd… <â¿â>

When metadata is being created (metadata import or file uploads), a step is applied where the names assigned to things (i.e. columns) is used to help suggest a usage.

In FM, the heuristics were primarily driven by explicit metadata in returned by a database (constraints), and perhaps the data type.

Looking at the actual data values themselves, can also provide additional metadata used by NLP etc.

https://www.ibm.com/docs/en/cognos-analytics/11.1.0?topic=metadata-data-sampling

I'm a long-time Framework Manager user that's slowly easing into the world of Data Modules. And for the most part I can get things to work, but now I'm trying to figure out "how" things work. What I want to know is "how" does Cognos do the metadata processing for sources like uploaded CSV files going into a Data Module? For example, it does a good job of picking out "identifiers", "facts", and "attributes", but how does Cognos know what "usage" values to assign to those various data items that are coming out of a generic source like CSV? What metadata criteria does it use for the assignments and at what stage in the process is the criteria applied?

I checked out the IBM Cognos Analytics v11.2 Metadata Modeling Guidelines, but it doesn't drill down into the stuff I'm looking for. So does anyone out there in Cognos Land have the info that will set my mind at ease… *^_^*

Thanks in advance, for any helpful insight.

Syd… <â¿â>

When metadata is being created (metadata import or file uploads), a step is applied where the names assigned to things (i.e. columns) is used to help suggest a usage.

In FM, the heuristics were primarily driven by explicit metadata in returned by a database (constraints), and perhaps the data type.

Looking at the actual data values themselves, can also provide additional metadata used by NLP etc.

https://www.ibm.com/docs/en/cognos-analytics/11.1.0?topic=metadata-data-sampling

I'm a long-time Framework Manager user that's slowly easing into the world of Data Modules. And for the most part I can get things to work, but now I'm trying to figure out "how" things work. What I want to know is "how" does Cognos do the metadata processing for sources like uploaded CSV files going into a Data Module? For example, it does a good job of picking out "identifiers", "facts", and "attributes", but how does Cognos know what "usage" values to assign to those various data items that are coming out of a generic source like CSV? What metadata criteria does it use for the assignments and at what stage in the process is the criteria applied?

I checked out the IBM Cognos Analytics v11.2 Metadata Modeling Guidelines, but it doesn't drill down into the stuff I'm looking for. So does anyone out there in Cognos Land have the info that will set my mind at ease… *^_^*

Thanks in advance, for any helpful insight.

Syd… <â¿â>