Skip to main content

Inter-Record Range Difference Processor

The Inter-Record Range Difference Processor (also called Range Difference Processor) transforms pairs of start/end columns into an internal representation that preserves their relationships. This is useful for temporal data, date ranges, and ordered pairs.

Description​

This processor handles data where pairs of variables represent a range or interval:

  • Start/end dates: start_date and end_date
  • Start/end times: start_time and end_time
  • Lower/upper bounds: min_value and max_value

The processor transforms these pairs into an internal representation that:

  1. Preserves the temporal or logical order (start < end)
  2. Maintains the distance between the two values
  3. Allows avatarization while keeping the structure intact
  4. Automatically sorts by the start variable

Use case: When your data contains time periods, events with duration, or ordered pairs that must maintain their relative ordering.

Concrete example​

Consider a dataset with event data:

person_id | event_name | start_date | end_date
1         | vacation   | 2024-01-15 | 2024-01-22
2         | project    | 2024-02-01 | 2024-03-15
3         | conference | 2024-03-20 | 2024-03-25

Problem without processor: During avatarization, start_date and end_date might be transformed independently, potentially creating impossible dates like end_date < start_date.

After applying Inter-Record Range Difference Processor:

Preprocessing step:
  - Records the original start_date values for each person_id
  - Calculates the duration: end_date - start_date
  - Sorts by start_date internally
  
During avatarization:
  - Dates are anonymized as a single unit (start_date and duration are preserved together)
  
Post-processing step:
  - Reconstructs end_date = avatarized_start_date + original_duration
  - Ensures end_date > start_date always holds

Result: Avatars maintain valid date ranges with preserved temporal logic.

How to Use it in Web​

In the web interface, the processor is not available yet. You can either use the Python API or process your data on your own before uploading it to the web interface.

How to use it in python​

Please refer to the Python documentation for usage details.

Key Points​

  • Order matters: Processors are applied in the order specified. If combining multiple processors, the start/end processor should typically be applied before others that modify time-based columns.
  • Multiple ranges: You can apply this processor to multiple start/end pairs by creating separate processor instances.
  • Non-date ranges: The processor also works with numeric ranges (e.g., min_score and max_score), as long as the start < end relationship is logical.