Getting Started with the RMLEditor
Pieter Heyvaert
Data Science Lab (Ghent University - iMinds)
What You Will Learn
- create mappings
- edit mappings
- run mappings
- load data sources and mappings
- export mappings
- suggested tasks
What You Will Learn
- create mappings
- edit mappings
- run mappings
- load data sources and mappings
- export mappings
- suggested tasks
Entity
An entity is something in the world, identified by a unique name (URI).
Anything can be an entity, including physical things, documents, abstract concepts, numbers and strings.
examples
- people
- companies
- buildings
Create an Entity
Entity
Blank Node
When no URI is provided for an entity, we call it a blank node.
Blank Node
Create a Blank Node
Information about an entity is represented by both attributes and relationships.
Examples of Attributes.
- "John Doe" (name of a person)
- "BE0596.342.234" (VAT number of a company)
Create an Attribute
Attribute
Relationships connect entities and their attributes.
example
- http://www.example.com/john_doe (entity)
- foaf:name (relationship, meaning "name of something/somebody")
- "John Doe" (attribute)
Relationships are also possible between entities
example
- http://www.example.com/john_doe (entity)
- dbpedia-owl:partner (relationship, meaning "partner of")
- http://www.example.com/jane_doe (entity)
Create Relationships
connect an entity and its attributes
However, more information is required to complete the mappings.
What You Will Learn
- create mappings
- edit mappings
- run mappings
- load data sources and mappings
- export mappings
- suggested tasks
Edit Mappings
entity
attribute
relationship
select = darker color
RDF Terms
- entity (URI)
- blank node (no URI)
- attribute
RDF Term Generation
- data extract
- templated data extract
- constant value
Entity
Entity type explains what the entity represents
Data extract or templated data extract
or constant value
Data extract
Source is the origin of the data
Column states the column of the data
Templated data extract
Source is the origin of the data
Template describes what happens
to the original data
Template
http://www.example.com/{ID}
| ID |
Transformation |
| 1 |
http://www.example.com/1 |
| 2 |
http://www.example.com/2 |
| 3 |
http://www.example.com/3 |
| 4 |
http://www.example.com/4 |
Constant value
Value is the constant value to use
for a resource, this has to be a valid URI
Blank Node
only optionally specify Entity type,
as we don't need to generate a URI.
Attribute
- allow to give more information about an entity (together with relationships)
- in most cases attributes are generated from a data extract or a templated data extract
Attribute
- data extract
- template data extract
- combine first name last name → {first_name} {last_name}
- constant value
- name of the company where employees work for, if not defined in data source
Attribute
same options as for an entity
Attribute type instead of entity type:
format of the data (e.g., string, integer)
state the language of the information
data is not a URI
Relationship
Constant is the most used option
when there is a relationship between two entities, as more information is required
Every event takes place on a specific location.
all events will be connected to all locations
events will only be connected to the location where they actually take place
Creating mappings using the schema-driven or model-driven approach
Mappings can become large graphs...
Levels of Detail
- 5 levels
- highest: maximize the details of the mapping
- lowest: maximize the overview
highest
All the information is visible.
highest
high
Attribute types are hidden.
high
moderate
Relationship details are hidden.
moderate
low
Attributes and entity types are hidden.
low
lowest
Blank nodes are hidden.
lowest
What You Will Learn
- create mappings
- edit mappings
- run mappings
- load data sources and mappings
- export mappings
- suggested tasks
Run Mappings
data sources + mapppings
= Linked Data
What You Will Learn
- create mappings
- edit mappings
- run mappings
- load data sources and mappings
- export mappings
- suggested tasks
Load Data Sources
View Loaded Data Sources
Load Mappings
What You Will Learn
- create mappings
- edit mappings
- run mappings
- load data sources and mappings
- export mappings
- suggested tasks
Export Mappings
to edit later
to run on other machine
What You Will Learn
- create mappings
- edit mappings
- run mappings
- load data sources and mappings
- export mappings
- suggested tasks
Suggested Tasks
- load data sources
- create (basic) mappings based on these data sources
- create entities and attributes
- define relationships
- run mappings to see results
- edit mappings where required
- go back to step 3 until happy
- export mappings
Actions
- Add entity/attribute: right click data or white space in modeling/center panel → click element
- Add relationship: select source of relationship → right click target of relationship → click Add Relation
- Delete entity/attribute/relationship: right click → click Delete
- Edit details of entity/attribute/relationship: select it → edit details in panel on the right
- Load data source: menu File → Open → Data Source → Click File or URI
- Load mappings: menu File → Open → Mapping
- Export mappings: menu File → Export → Mapping
