Skip to content

Latest commit

 

History

History
141 lines (103 loc) · 5.19 KB

tutorial.md

File metadata and controls

141 lines (103 loc) · 5.19 KB
Raw

Tutorial

Motivation

The main use-case for tripsu is to be used in combination with other CLI tools up- and downstream via piping. Let us assume that we're running a SPARQL query on a large graph and we would like to pseudonymize some of the triples. This is how the flow should look like:

curl <sparql-query> | tripsu pseudo -x index.nt -r rules.yaml > pseudo.nt

For this flow to stream data instead of loading everything into memory, note that an indexing step is required to allow the pseudonymization to run on a stream without loading the graph into memory.

Example

There are three possible ways to pseudonymize RDF triples:

  1. Pseudonymize the URI of nodes with rdf:type.
  2. Pseudonymize values for specific subject-predicate combinations.
  3. Pseudonymize any value for a given predicate.

By combining these, can process an RDF file with sensitive information:

<http://example.org/Alice> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> .
<http://example.org/Alice> <http://xmlns.com/foaf/0.1/holdsAccount> <http://example.org/Alice-Bank-Account> .
<http://example.org/Alice-Bank-Account> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/OnlineAccount> .
<http://example.org/Alice-Bank-Account> <http://schema.org/name> "my_account32" .
<http://example.org/Alice-Bank-Account> <http://schema.org/accessCode> "secret-123" .
<http://example.org/Alice> <http://schema.org/name> "Alice" .
<http://example.org/Bank> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Organization> .
<http://example.org/Bank> <http://schema.org/name> "Bank" .

into a pseudonymized file where the sensitive information such as people's names, personal and secret information is hashed to protect privacy:

<http://example.org/af321bbc> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> .
<http://example.org/af321bbc> <http://xmlns.com/foaf/0.1/holdsAccount> <http://example.org/bs2313bc> .
<http://example.org/bs2313bc> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/OnlineAccount> .
<http://example.org/bs2313bc> <http://schema.org/name> "pp54r32" .
<http://example.org/bs2313bc> <http://schema.org/accessCode> "asfnd223" .
<http://example.org/af321bbc> <http://schema.org/name> "af321bbc" .
<http://example.org/Bank> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Organization> .
<http://example.org/Bank> <http://schema.org/name> "Bank" .

The next subsections break down each of the three pseudonymization approaches to better understand how they operate.

1. Pseudonymize the URI of nodes with rdf:type

Given the following config:

nodes:
  of_type:
  - "<http://xmlns.com/foaf/0.1/Person>"

The goal is to pseudonymize all instaces of rdf:type Person. The following input file:

<http://example.org/Alice> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> .

Would become:

<http://example.org/af321bbc> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> .

2. Pseudonymize values for specific subject-predicate combinations

Given the following config:

objects:
  on_type_predicate:
    "<http://xmlns.com/foaf/0.1/Person>":
    - "<http://schema.org/name>"

The goal is to pseudonymize only the instances of names when they're associated to Person. The following input file:

<http://example.org/Alice> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> .
<http://example.org/Alice> <http://schema.org/name> "Alice" .
<http://example.org/Bank> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Organization> .
<http://example.org/Bank> <http://schema.org/name> "Bank" .

Would become:

<http://example.org/Alice> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> .
<http://example.org/Alice> <http://schema.org/name> "af321bbc" .
<http://example.org/Bank> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Organization> .
<http://example.org/Bank> <http://schema.org/name> "Bank" .

3. Pseudonymize any value for a given predicate

Given the following config:

objects:
  on_predicate:
  - "<http://schema.org/name>"

The goal is to pseudonymize any values associated to name. The following input file:

<http://example.org/Alice> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> .
<http://example.org/Alice> <http://schema.org/name> "Alice" .
<http://example.org/Bank> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Organization> .
<http://example.org/Bank> <http://schema.org/name> "Bank" .

Would become:

<http://example.org/Alice> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> .
<http://example.org/Alice> <http://schema.org/name> "af321bbc" .
<http://example.org/Bank> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Organization> .
<http://example.org/Bank> <http://schema.org/name> "38a3dd71" .