DBpedia Lookup is a web service that can be used to look up DBpedia URIs by related keywords. Related means that either the label of a resource matches, or an anchor text that was frequently used in Wikipedia to refer to a specific resource matches (for example the resource http://dbpedia.org/resource/United_States can be looked up by the string "USA"). The results are ranked by the number of inlinks pointing from other Wikipedia pages at a result page.

Two APIs are offered: Keyword Search and Prefix Search. A hosted version of the Lookup service is available on the DBpedia server infrastructure.

The Keyword Search API can be used to find related DBpedia resources for a given string. The string may consist of a single or multiple words.

Example: Places that have the related keyword "berlin"

http://lookup.dbpedia.org/api/search/KeywordSearch?QueryClass=place&QueryString=berlin

The Prefix Search API can be used to implement autocomplete input boxes. For a given partial keyword like berl the API returns URIs of related DBpedia resources like http://dbpedia.org/resource/Berlin.

Example: Top five resources for which a keyword starts with "berl"

http://lookup.dbpedia.org/api/search/PrefixSearch?QueryClass=&MaxHits=5&QueryString=berl

The query parameters accepted by the endpoints are

• QueryString: a string for which a DBpedia URI should be found.
• QueryClass: a DBpedia class from the Ontology that the results should have (for owl#Thing and untyped resource, leave this parameter empty).
• MaxHits: the maximum number of returned results (default: 5)

By default all data is returned as XML, the service also retuns JSON to any request including the Accept: application/json header.

DBpedia Lookup depends on the core of the DBpedia extraction framework, which is not available in a public Maven repo at the moment. Java 7 is required to compile it.

git clone git://github.com/dbpedia/extraction-framework.git
cd extraction-framework
git checkout DBpedia_3.8
mvn clean install

git clone git://github.com/dbpedia/lookup.git
cd lookup
mvn clean install

wget http://spotlight.dbpedia.org/download/dbpedia-lookup-index-3.8.tgz
tar xzvf dbpedia-lookup-index-3.8.tgz

./run Server dbpedia-lookup-index-3.8

The server should now be running at http://localhost:1111

Rebuilding an index is usually not required, if you only intend on running a local mirror of the service you can donwload a prebuilt index as outlined above.

To re-build the index you will require

• DBpedia datasets
• pignlproc output from nerd-stats.pig
• Unix

from http://downloads.dbpedia.org/current/en/

• redirects_en.nt
• short_abstracts_en.nt
• instance_types_en.nt
• article_categories_en.nt

This is necessary because indexing in sorted order is significantly faster.

  cat instance_types_en.nt  \
      short_abstracts_en.nt \
      article_categories_en.nt | sort >all_dbpedia_data.nt

Redirects are not indexed, but they are excluded as targets of lookup.

The indexer has to be run twice:

1. with the DBpedia data

    ./run Indexer lookup_index_dir redirects_en.nt all_dbpedia_data.nt
   

2. with the pignlproc data

    ./run Indexer lookup_index_dir redirects_en.nt nerd_stats_output.tsv
   

The best way to get support or give feedback on the Lookup project is via the DBpedia discussion mailing list. More technical queries about the code base should be directed to the DBpedia developers mailing list.

The DBpedia wiki also has useful information on the project.

• Pablo Mendes @pablomendes
• Max Jakob @maxjakob
• Matt Haynes @matth