Let me take this opportunity to explain our product called J-Gate which is a bibliographic database acting as a discovery product for Scholars, Researchers, Scientist community and Post graduate students.
J-Gate indexes 52000+ journals along with thousands of Theses, Conferences etc. We have clients scattered with category from Aeronautics to Zoology.
Since ours is a bibiliographic & scholarly database , we need to provide powerful search tools for narrowing results, where users should be able to more quickly find the information, they need.
We are looking for a solution in annotating the documents to generate keywords for entire documents of J-Gate( ~70 million plus) in bulk.
How does DBpedia helps us in achieving this?
What are the skills required to use the API available in DBpedia?,
How to use the Open knowledge graph of DBpedia?
Do you have any services offered for our kind of requirement?
Can you share the ontology being used for annotating the document in DBpedia? for our study purpose?
Dear @shylaja,
Please have a look at our annotation tool spotlight: https://demo.dbpedia-spotlight.org/ It has a free API and also a docker to self-deploy at J-Gate. It also comes in 20-30 languages and you could add more languages.
Spotlight gives you entity links for your text. You can do the following:
- the entity URLs connect it to DBpedia, i.e. you can query more information (see https://dbpedia.org/resource/J-Gate) everything that is in Wikipedia.
- even more: Proceedings of the Posters and Demos Track of the 13th International Conference on Semantic Systems -- 11th – 14th of September 2017. Amsterdam, The Netherlands was an annotated HTML proceeding which could be queried with SPARQL
see Linked Data Access - DBpedia Organization or SPARQL over Online Databases - DBpedia Organization
Default skills are required, i.e. HTTP requests and SPARQL queries.
see links above
DBpedia Core (English) has 7 million entities, which are exactly the same as Wikipedia pages. We also have DBpedia in 140 languages. These are used to annotate in spotlight. This creates a connection to the knowledge graph, where you can find more information. See the HTML interface of J-Gate: About: J-Gate it has e.g. this info:
dbp:type dbr:Bibliographic_database
is dbo:product of dbr:Informatics_India_Ltd
dct:subject
dbc:Academic_publishing
dbc:Electronic_publishing
and many more. There is also an ontology: Ontology
We also have an extraction of Wikidata, that cleans wikidata and makes it more usable and compatible with DBpedia: databus.dbpedia.org/dbpedia/wikidata