To provide users with optimal capabilities for data analysis and evaluation, the cloud version of Qlik Sense is continuously being developed. As part of this effort, the Table Recipe was introduced in the past and has recently been expanded – including with a feature to optimize data quality.
The Table Recipe in Qlik Sense has recently gained a new feature that expands its capabilities and makes it easier to work with. To this end, the Table Recipe has been equipped with new native data quality functions. These help users identify, visualize, and correct invalid or inconsistent data. The new functions make it easier to prepare cleaned, reliable datasets for analysis or machine learning. There is no need to leave the code-free interface, which is structured like a spreadsheet. The new features include:
- Detection of data and semantic types: Each column is automatically assigned a data type based on the data it contains. Data types include native types such as integer, date, and text, as well as semantic types such as predefined (email, phone number, ZIP code, country code, etc.) or user-defined (dictionary-based, pattern-based, or composite). The assigned data type can be manually adjusted at any time via the column menu.
- Column quality bar: Each column header displays a quality bar that shows the percentage of valid values (green), empty or null values (black), and invalid values (red).
- Cell-level indicators for invalid values: Cells containing invalid values are marked in red.
- Validity-based filtering: Clicking any segment of a column’s quality bar allows you to instantly filter rows by quality status (valid, invalid, or empty/zero). You can also filter using the column menu or the filter dialog box.
- Cleanse and contextual correction suggestions: The “Clean” function allows you to replace invalid, zero, or empty values in a column. The table recipe also displays relevant correction suggestions—such as replace, delete, or fill—based on detected invalid values, as well as suggestions for conversion functions based on the detected data type.
