Aim

Provide tools and in some cases usage scenarios for working with multilingual linked data sources. That may include e.g. typical queries for linked data sources to get translations for given units, tools for entities discovery, term search etc.

Approach

Each entry should provide:

1. Name of data source or tool
2. URI to access data source as sparql endpoint, or download of data
3. Queries + query description

DBpedia

Sparlq endpoint:

1. http://dbpedia.org/snorql
2. http://dbpedia.org/sparql

Example query:

SELECT ?o ?o2
WHERE {
  <http://dbpedia.org/resource/Eye> owl:sameAs ?o.
  FILTER regex(?o,'http://(.*).dbpedia')
  BIND(replace(str(?o),'(http://)(.*)(.dbpedia.org/resource/)(.*)','$2: $4','i') AS ?o2)
}

Query description: for resource http://dbpedia.org/resource/Eye gather links via owl:sameAs and filtering links that are mostly used for cross-lingual references. Then create a list of items including language identifier + translation.

Data sources documented in the BPMLOD group

Work for such queries is being done in the BPMLOD CG, see e.g.:

• http://bpmlod.github.io/report/multilingual-dictionaries/BabelNet/
• http://bpmlod.github.io/report/multilingual-dictionaries/BabelNet/#data-querying

One may consider joining above CG to avoid cross-posting with the ITS IG.

Dbnary

• Main page see http://kaiko.getalp.org/about-dbnary/
• Example queries: see http://kaiko.getalp.org/about-dbnary/?page_id=13
• SPARQL endpoint: http://kaiko.getalp.org/sparql

Further tools and services

tbd: create sections for:

• Yahoo Content Analysis
• Open Calais
• AlchemyAPI