employment

Scottish Government - Employment data example

This example aims to demonstrate how table2qb can be used to convert a sample of the Scottish government's employment data from CSV to RDF.

Directory overview

There are 2 folders in this directory:

1. csv folder contains the inputs:

   1.1 reference data:

   • components: components.csv
   • codelists: gender.csv and units.csv

   1.2 observation data:

   • input.csv

2. ttl folder contains the outputs:

   2.1 reference data:

   • components: components.ttl
   • codelists: gender.ttl and measurement-units.ttl

   2.2 observation data:

   • cube.ttl

Cube structure

The observations data indicates the required structure of the cube. This cube is a 'measure type' cube with an explicit qb:measureType dimension indicating the measure corresponding to each row. There is a single measure Count referenced by the Measure Type column. There are four dimension columns (including the qb:measureType column): Geography, Quarter, Gender and Measure Type. The Value column contains the observed Count for each observation, and the Unit column is an attribute of the measure which indicates how to interpret the observed value.

Components

The Geography and Quarter dimensions use existing vocabularies to define the reference area and period for the observation. The Measure Type dimension is also defined by data cube specification. This leaves only the Gender dimension and Count measure components which need to be defined. These components are specified within components.csv and generated by the components-pipeline:

table2qb exec components-pipeline --input-csv examples/employment/csv/components.csv --base-uri http://statistics.gov.scot/ --output-file examples/employment/ttl/components.ttl

Codelists

The gender dimension defined within components.csv has an associated codelist. The values of this codelist are defined in gender.csv and can be generated using codelist-pipeline:

table2qb exec codelist-pipeline --codelist-csv examples/employment/csv/gender.csv --codelist-name "Gender" --codelist-slug "gender" --base-uri http://statistics.gov.scot/ --output-file examples/employment/ttl/gender.ttl

Note the URI of the codelist is generated from the base-uri and codelist-slug parameters specified above into a URI of the form {base-uri}/def/concept-scheme/{codelist-slug}. The resulting URI of http://statistics.gov.scot/def/concept-scheme/gender should match the Codelist URI specified for the Gender dimension within components.csv.

The units of the measure also have an associated codelist defined within units.csv:

table2qb exec codelist-pipeline --codelist-csv examples/employment/csv/units.csv --codelist-name "Measurement Units" --codelist-slug "measurement-units" --base-uri http://statistics.gov.scot/ --output-file examples/employment/ttl/measurement-units.ttl

Cube

Once the components and codelists have been specified, the cube can be generated from the observation data. The configuration for the input columns is specified in columns.csv. This defines the four dimension columns, the Count measure and Unit attribute as well as the Value column. The Value column is identified as containing the measure value by its empty component_attachment. The value_template of the Gender dimension should produce URIs which match the members of the Gender concept scheme generated by the codelist-pipeline above. Codelist members are identified by URIs of the form {base-uri}/def/concept-scheme/{codelist-slug}/{notation} where notation is defined within the codelist CSV file. The value_template of http://statistics.gov.scot/def/concept/gender/{gender} for the Gender column matches this format. The values of the Gender column within the observation data are upper-case i.e. Female, Male, All. These values must be transformed before being substituted into the value URIs. The slugize value_transformation specified for the Gender column performs this transformation for the values in the input observation data. The cube is then generated with:

table2qb exec cube-pipeline --input-csv examples/employment/csv/input.csv --dataset-name "Employment" --dataset-slug "employment" --column-config examples/employment/columns.csv --base-uri http://statistics.gov.scot/ --output-file examples/employment/ttl/cube.ttl

Further processing

Given the commands specified above, the resulting data cube will be defined within the components.ttl, gender.ttl, measurement-units.ttl and cube.ttl files.