Enhancing Library Discovery with
Linked Open Data

IGeLU/ELUNA Show & Tell, August 1, 2016

Steve Meyer, Data Strategist, UW-Madison Libraries

A Little Context

library.wisc.edu/experiments/linked-data

Sample URLs

Catalog Enhanced with Linked Data:
https://search-ld.library.wisc.edu/

• Gertrude Stein on Picasso
• Kind of Blue

Info Card

Example In Context

How does this work?
Data Flow

Systems in Play

Bibliographic Identity Crawl

Step 2.1: Resolving VIAF URI
(and corresponding data!)

http://viaf.org/viaf/sourceID/LC|n78086005

redirects to

http://viaf.org/viaf/15873

and then to

http://viaf.org/viaf/15873/

Which VIAF Graph Entity?

<rdf:RDF>
  <rdf:Description rdf:about="http://viaf.org/viaf/15873/"/>
  <rdf:Description rdf:about="http://viaf.org/viaf/15873"/>
  <rdf:Description rdf:about="http://viaf.org/viaf/sourceID/BAV%7CADV12539089#skos:Concept"/>
  <rdf:Description rdf:about="http://viaf.org/viaf/sourceID/LC%7Cn++78086005#skos:Concept"/>
  ...
</rdf:RDF>

Ask the Graph, Use What is Known

Already known:

{
  "@id":"http://id.loc.gov/authorities/names/n78086005",
  "label":"Picasso, Pablo, 1881-1973.",
  "sameAs":"http://viaf.org/viaf/sourceID/LC|n78086005"
}

Query:

sameas_uri  = RDF::URI.new("http://schema.org/sameAs")
creator_uri = RDF::URI.new("http://id.loc.gov/authorities/names/n78086005")
graph.query(predicate: sameas_uri, object: creator_uri)

VIAF Entity Resolved

The query produces the following RDF triple.

Step 2.2: Find VIAF Person Entity Links

Example: Find the DBpedia URI

sameas_uri     = RDF::URI.new("http://schema.org/sameAs")
entity_subject = RDF::URI.new("http://viaf.org/viaf/15873")
other_entities = graph.query(subject: entity_subject, predicate: sameas_uri)
other_entities.select {|s| s.object.to_s.match('http://dbpedia.org/resource')}.first

Resulting Triple

Rinse, Repeat...

1. <http://vocab.getty.edu/ulan/500009666-agent>
2. <http://www.wikidata.org/entity/Q5593>
3. <http://dbpedia.org/resource/Pablo_Picasso>

Crawl or Query?

Sample SPARQL

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX owl: <http://www.w3.org/2002/07/owl#>
PREFIX foaf: <http://xmlns.com/foaf/0.1/>
PREFIX dbo: <http://dbpedia.org/ontology/>

SELECT ?influenced ?influencedGivenName ?influencedSurname ?influencedSameAs
WHERE {
  { 
    <http://dbpedia.org/resource/Pablo_Picasso> dbo:influenced ?influenced . 
  }
  UNION
  {
    ?influenced dbo:influencedBy <http://dbpedia.org/resource/Pablo_Picasso> .
  }
  ?influenced foaf:givenName ?influencedGivenName .
  ?influenced foaf:surname ?influencedSurname .
  OPTIONAL { 
    ?influenced owl:sameAs ?influencedSameAs . 
    FILTER regex(STR(?influencedSameAs), "viaf.org").
  }
}

Crawl vs. Query Comparison

Picasso's influence according to DBpedia

• SPARQL query: 1 HTTP request
• HTTP Crawl: 54 HTTP requests
  • Initial HTTP request for primary subject
  • 1 HTTP request per influence related entity
    Picasso was an influence upon 53 other entities in DBpedia

How does this work?
Code

BibCard

Design Goals

Environment / Context

Library Discovery ⊃ Web Application

Technical Considerations

1. Maximize Perceived Speed via
   • Caching Mechanisms
   • Asynchronous Enhancement
2. Programming Model: Object Oriented

BibCard Core Responsibilities

1. Fetch data that can be cached
2. Instantiate Person object

1) Fetch Cacheable Data

# Given an LCNAF or VIAF URI,
# Resolve the related URIs,
# Query for relevant data points,
# Return a micrograph
data = BibCard.person_data("http://id.loc.gov/authorities/names/n78086005")

puts data

# <http://viaf.org/viaf/15873> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://schema.org/Person> .
# <http://viaf.org/viaf/15873> <http://schema.org/deathDate> "1973-04-09" .
# <http://viaf.org/viaf/15873> <http://schema.org/sameAs> <http://id.loc.gov/authorities/names/n78086005> .
# ...

Micrograph

A small heterogeneous collection of RDF triples from multiple sources

<http://viaf.org/viaf/85312226> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://schema.org/Person> .
<http://viaf.org/viaf/85312226> <http://schema.org/name> "Tim Berners-Lee"@en .
<http://viaf.org/viaf/85312226> <http://schema.org/birthDate> "1955-06-08" .
<http://viaf.org/viaf/85312226> <http://schema.org/sameAs> <http://dbpedia.org/resource/Tim_Berners-Lee> .
<http://viaf.org/viaf/85312226> <http://schema.org/sameAs> <http://www.wikidata.org/entity/Q80> .
<http://dbpedia.org/resource/Tim_Berners-Lee> <http://dbpedia.org/ontology/influencedBy> <http://dbpedia.org/resource/Paul_Otlet> .
<http://dbpedia.org/resource/Tim_Berners-Lee> <http://dbpedia.org/ontology/influencedBy> <http://dbpedia.org/resource/Jon_Postel> .
<http://dbpedia.org/resource/Tim_Berners-Lee> <http://dbpedia.org/ontology/abstract> "Professor Sir Timothy John Berners-Lee, OM, KBE, FRS, FREng, FRSA, DFBCS (born 8 June 1955), also known as TimBL, is an English computer scientist, best known as the inventor of the World Wide Web. He made a proposal for an information management system in March 1989, and he implemented the first successful communication between a Hypertext Transfer Protocol (HTTP) client and server via the Internet sometime around mid-November of that same year.Berners-Lee is the director of the World Wide Web Consortium (W3C), which oversees the Web's continued development. He is also the founder of the World Wide Web Foundation, and is a senior researcher and holder of the Founders Chair at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL).He is a director of the Web Science Research Initiative (WSRI), and a member of the advisory board of the MIT Center for Collective Intelligence.  In 2011 he was named as a member the Board of Trustees of the Ford Foundation.In 2004, Berners-Lee was knighted by Queen Elizabeth II for his pioneering work. In April 2009, he was elected a foreign associate of the United States National Academy of Sciences. He was honoured as the \"Inventor of the World Wide Web\" during the 2012 Summer Olympics opening ceremony, in which he appeared in person, working with a vintage NeXT Computer at the London Olympic Stadium. He tweeted \"This is for everyone\", which instantly was spelled out in LCD lights attached to the chairs of the 80,000 people in the audience." .
<http://dbpedia.org/resource/Paul_Otlet> <http://xmlns.com/foaf/0.1/givenName> "Paul" .
<http://dbpedia.org/resource/Paul_Otlet> <http://xmlns.com/foaf/0.1/surname> "Otlet" .
<http://dbpedia.org/resource/Paul_Otlet> <http://www.w3.org/2002/07/owl#sameAs> "http://viaf.org/viaf/54277595" .
<http://dbpedia.org/resource/Jon_Postel> <http://xmlns.com/foaf/0.1/givenName> "Jon" .
<http://dbpedia.org/resource/Jon_Postel> <http://xmlns.com/foaf/0.1/surname> "Postel" .
<http://www.wikidata.org/entity/Q80> <http://www.wikidata.org/prop/direct/P69> <http://www.wikidata.org/entity/Q73094> .
<http://www.wikidata.org/entity/Q80> <http://www.wikidata.org/prop/direct/P69> <http://www.wikidata.org/entity/Q5369138> .
<http://www.wikidata.org/entity/Q80> <http://www.wikidata.org/prop/P69> <http://www.wikidata.org/entity/statement/q80-166977F2-448C-48BB-B4B1-C14A3714DE82> .
<http://www.wikidata.org/entity/Q80> <http://www.wikidata.org/prop/P69> <http://www.wikidata.org/entity/statement/Q80-E892C384-4E42-4E6E-9A3D-6E6FE3201954> .
<http://www.wikidata.org/entity/Q80> <http://schema.org/description> "Web developer" .
<http://www.wikidata.org/entity/Q80> <http://www.wikidata.org/prop/direct/P937> <http://www.wikidata.org/entity/Q42944> .
<http://www.wikidata.org/entity/Q73094> <http://www.w3.org/2000/01/rdf-schema#label> "The Queen's College" .
<http://www.wikidata.org/entity/statement/q80-166977F2-448C-48BB-B4B1-C14A3714DE82> <http://www.wikidata.org/prop/statement/P69> <http://www.wikidata.org/entity/Q73094> .
<http://www.wikidata.org/entity/Q5369138> <http://www.w3.org/2000/01/rdf-schema#label> "Emanuel School" .
<http://www.wikidata.org/entity/statement/Q80-E892C384-4E42-4E6E-9A3D-6E6FE3201954> <http://www.wikidata.org/prop/statement/P69> <http://www.wikidata.org/entity/Q5369138> .
<http://www.wikidata.org/entity/Q42944> <http://www.w3.org/2000/01/rdf-schema#label> "European Organization for Nuclear Research" .

2) Instantiate Code Objects

Code Flow: Instantiating Person Object

1. Get micrograph for URI
   use web app framework's cache
2. Load a Spira repository from the raw data
3. Using the repository graph, find VIAF URI
4. Instantiate BibCard::Person with VIAF URI

Using Spira Objects in View Code

Spira Model Objects

Intended to look similar to Rails ActiveRecord associations

URI Hack #2: Finding URIs by String Match

Request Timeouts

Request Timeout Consequences

Functionality can only be a progressive enhancement
(Linked Data info cards are additive, not core data)

• Caching should be neither too short, nor too long
• Content added to the page via asynchronous requests

Image Credits