linkedin icon twitter icon facebook icon

Private Area

headline

Do you need some more informations?

Fill our contact form >

Reliable data intelligence

At Nextbit we achieve reliability on data intelligence through diagnostics and a set of reporting tools that monitor the data quality over time.

Diagnostic>

Within our diagnostic framework we use three levels of Diagnostic layers:

  • Level 0 – monitors the structures of tables
  • Level 1 – monitors the value of each record variable against the valid domain
  • Level 2 – uses business and logical rules for record level atomic data consistency and model validation

These three levels form our diagnostic framework which is deployed and executed with every data update process.

data process

Reporting>

Automated Reporting is an important step on achieving Reliable data intelligence. The contents of the automated reporting set up by Nextbit varies from project to project depending on needs and project scope. Below is a sample of some of items that we consider strategic to a data intelligence projects and fundamental towards proactive communication to analysts and user on the data available for analysis and modeling. Nextbit provides its clients with programs and code to readily deploy the following:

  • Automated Variable Statistics Report (Insightful for modeling purposes)
    • Percentage of missing
    • Statistics on continuous numeric variables
    • Statistics on discrete (classification / ordinal) variables
    • Outliers indicator
  • Auto-Updated Data Dictionary (Necessary for up-to-date data analysis and modeling)
    • An automatically updated data dictionary with variable and transformation names, descriptions and valid domains.
  • Parameters and criteria Reporting (Insightful for modeling purposes and model maintenance)
    • Variable selection for inclusion/exclusion in models
    • Rules/Criteria for exclusion (based on variable values)
    • Limit controls and tolerance levels on historical data values
  • System Reporting (Necessary for up-to-date data analysis)
    • Date and time of most recent update of indicator/variable
    • Diagnostic rules results on single variable/KPI