KAnonimity Configuration

To initialise and run the component two configs are used - general_config.ini and kanonimity.ini. In general_config.ini all paths to the corresponding data objects shall be specified. Example:

[Paths.Silver]
...
# Present Population UC input data object
estimated_present_population_zone_silver = ${Paths:silver_dir}/estimated_present_population_zone

# Usual Environment UC input data objects
estimated_aggregated_usual_environments_zone_silver = ${Paths:silver_dir}/estimated_aggregated_usual_environment_zone
...

[Paths.Gold]
# Present Population UC output data object
kanonimity_present_population_zone_gold = ${Paths:gold_dir}/kanonimity_present_population_zone

# Usual Environment UC output data object
kanonimity_aggregated_usual_environments_zone_gold = ${Paths:gold_dir}/kanonimity_aggregated_usual_environment_zone
...

General section

The specific component configuration file, kanonimity.ini, has three different sections. The general section [KAnonimity] is used to indicate on what UC outputs the k-anonimity process should be executed. Its expected parameters are: - present_population_execution: boolean, it indicates whether the k-anonimity process should be executed on the zone-aggregated present population output. Example: True. - usual_environment_execution: boolean, it indicates whether the k-anonimity process should be executed on the zone-aggregated usual environment output. Example: True.

## Present Population section If the configuration parameter present_population_execution has been set to True, the component will read the section related to Present Population, [KAnonimity.PresentPopulationEstimation]. The expected parameters in this section are as follows: - clear_destination_directory: boolean, whether to delete all previous results in the output directory before running the component. - k: integer, it indicates the threshold value to select what records should have k-anonimity applied to them, that is, all records that have a population value strictly lower than k. Example: 5. - anonimity_type: string, it indicates what k-anonimity metholodogy to employ. If it is set to delete, it will delete all records that have a population value strictly lower than k. If it is set to obfuscate, it will replace all population values strictly lower than k by the negative value -1. Example: delete. - zoning_dataset_id: string, ID of the zones dataset that was previously used to aggregate the data from grid level to zone level. This identifies what specific present population output data should be processed by this component. Example: nuts. - hierarchical_levels: comma-separated list of positive integers, they are the hierarchical levels of the zones dataset used that should be processed by this component. If the zones dataset was not hierarchical, this should just take the value 1. Example: 1,2,3. - start_date: string, in YYYY-MM-DD format, indicates the starting date (inclusive) for which the present population data should be processed by this component. Example: 2023-01-01. - end_date: string, in YYYY-MM-DD format, indicates the ending date (inclusive) for which the present population data should be processed by this component. Example: 2023-01-01.

## Usual Environment section If the configuration parameter usual_environment_execution has been set to True, the component will read the section related to Usual Environment, [KAnonimity.UsualEnvironmentAggregation]. The expected parameters in this section are as follows: - clear_destination_directory: boolean, whether to delete all previous results in the output directory before running the component. - k: integer, it indicates the threshold value to select what records should have k-anonimity applied to them, that is, all records that have a weighted device count value strictly lower than k. Example: 5. - anonimity_type: string, it indicates what k-anonimity metholodogy to employ. If it is set to delete, it will delete all records that have a weighted device count value strictly lower than k. If it is set to obfuscate, it will replace all weighted device count values strictly lower than k by the negative value -1. Example: delete. - zoning_dataset_id: string, ID of the zones dataset that was previously used to aggregate the data from grid level to zone level. This identifies what specific usual environment output data should be processed by this component. Example: nuts. - hierarchical_levels: comma-separated list of positive integers, they are the hierarchical levels of the zones dataset used that should be processed by this component. If the zones dataset was not hierarchical, this should just take the value 1. Example: 1,2,3. - labels: comma-separated list of strings, they indicate what usual environment labels should be processed by this component. Allowed values are home, work, ue. Example: ue, home, work. - start_month: string, in YYYY-MM format, it indicates the first month of the period used to compute a specific usual environment dataset that the user desires to process through this component. Example: 2023-01. - end_month: string, in YYYY-MM format, it indicates the last month of the period used to compute a specific usual environment dataset that the user desires to process through this component. Example: 2023-03. - season: string, value of the season of the usual environment dataset that the user desires to prcess through this component. Allowed values are: all, spring, summer, autumn, winter. Example: all.

Configuration example

[Spark]
session_name = KAnonimity

[KAnonimity]
present_population_execution = True
usual_environment_execution = True


[KAnonimity.PresentPopulationEstimation]
clear_destination_directory = True
anonimity_type = obfuscate  # `obfuscate`, `delete`
k = 5

# Target PresentPopulationEstimation dataset
zoning_dataset_id = nuts  # ID of the zoning dataset
hierarchical_levels = 1,2,3,4  # Hierarchical level(s) of the zoning dataset. Comma-separated list
start_date = 2023-01-01  # start date (inclusive)
end_date = 2023-01-01  # end date (inclusive)

[KAnonimity.UsualEnvironmentAggregation]
clear_destination_directory = True
anonimity_type = obfuscate  # `obfuscate`, `delete`
k = 5

# Target UsualEnvironmentAggregation dataset
zoning_dataset_id = nuts  # ID of the zoning dataset
hierarchical_levels = 1,2,3,4  # Hierarchical level(s) of the zoning dataset. Comma-separated list
labels = ue, home, work  # Allowed values: `ue`, `home`, `work`. Comma-separated list
start_month = 2023-01  # Start month (inclusive)
end_month = 2023-03  # End month (inclusive)
season = all  # Allowed values: `all`, `spring`, `summer`, `autumn`, `winter`.