Skip to main content

A framework to convert natural language to database queries.

Project description

  __ _ _   _  ___ _ __  _   _
 / _` | | | |/ _ \ '_ \| | | |
| (_| | |_| |  __/ |_) | |_| |
 \__, |\__,_|\___| .__/ \__, |
    |_|          |_|    |___/

What’s quepy?

Quepy is a python framework to transform natural language questions to queries in a database query language. It can be easily customized to different kinds of questions in natural language and database queries. So, with little coding you can build your own system for natural language access to your database.

Currently Quepy provides support for Sparql and MQL query languages. We plan to extended it to other database query languages.

An example

To illustrate what can you do with quepy, we included an example application to access DBpedia contents via their sparql endpoint.

You can try the example online here: Online demo

Or, you can try the example yourself by doing:

python examples/dbpedia/main.py "Who is Tom Cruise?"

And it will output something like this:

SELECT DISTINCT ?x1 WHERE {
    ?x0 rdf:type foaf:Person.
    ?x0 rdfs:label "Tom Cruise"@en.
    ?x0 rdfs:comment ?x1.
}

Thomas Cruise Mapother IV, widely known as Tom Cruise, is an...

The transformation from natural language to sparql is done by first using a special form of regular expressions:

person_name = Group(Plus(Pos("NNP")), "person_name")
regex = Lemma("who") + Lemma("be") + person_name + Question(Pos("."))

And then using and a convenient way to express semantic relations:

person = IsPerson() + HasKeyword(person_name)
definition = DefinitionOf(person)

The rest of the transformation is handled automatically by the framework to finally produce this sparql:

SELECT DISTINCT ?x1 WHERE {
    ?x0 rdf:type foaf:Person.
    ?x0 rdfs:label "Tom Cruise"@en.
    ?x0 rdfs:comment ?x1.
}

Using a very similar procedure you could generate and MQL query for the same question obtaining:

[{
    "/common/topic/description": [{}],
    "/type/object/name": "Tom Cruise",
    "/type/object/type": "/people/person"
}]

Installation

You need to have installed numpy. Other than that, you can just type:

pip install quepy

You can get more details on the installation here:

http://quepy.readthedocs.org/en/latest/installation.html

Learn more

You can find a tutorial here:

http://quepy.readthedocs.org/en/latest/tutorial.html

And the full documentation here:

http://quepy.readthedocs.org/

Join our mailing list

Contribute!

Want to help develop quepy? Welcome aboard! Find us in

  • #quepy at freenode

  • quepydev at librelist.com

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

quepy-0.2.tar.gz (42.3 kB view details)

Uploaded Source

File details

Details for the file quepy-0.2.tar.gz.

File metadata

  • Download URL: quepy-0.2.tar.gz
  • Upload date:
  • Size: 42.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for quepy-0.2.tar.gz
Algorithm Hash digest
SHA256 72b563c5f181935f66c9f3b6105400cfbe382f0821c60a119445374dc16cec52
MD5 093bebdf2f09535311f7bd6b0da72d78
BLAKE2b-256 e9bc42d951d3f4cb95186c7be736f493a75852f21e6d11484927963a09c16514

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page