Skip to main content

Natural language understanding system

Project description

Montague

Build Status

An experimental natural-language understanding system.

Unlike conventional, machine-learning NLU systems, Montague uses formal semantics to interpret English sentences.

When Montague reads a sentence, it looks up each word in its lexicon to assign it a meaning in its logical language (an extension of first-order logic). Then, it recursively combines the words into phrases and computes the meaning of each phrase. When the sentence has been translated into a logical formula, Montague evaluates the formula against its model of the world to determine its truth value.

Installation

$ pip3 install montague-nlu

Alternatively, you can install a development version with setup.py:

$ git clone https://github.com/iafisher/montague.git
$ python3 setup.py develop --user

Once you do so, you can play with Montague's command-line interface:

$ montague

Limitations

The Montague system is still in early beta and suffers from many limitations.

  • Montague has no knowledge of syntax. Its only criterion for grouping two phrases is whether they are linearly adjacent and whether their types are compatible. This leads Montague to interpret nonsense sentences like "Every good is child."
  • Montague will fail to interpret a sentence if it contains a word that is not in its lexicon.
  • Important modules of formal semantics, like plurality, tense, aspect, theta roles, intensionality, and indexicals, have yet to be implemented.

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

montague-nlu-0.1.6.tar.gz (12.8 kB view details)

Uploaded Source

Built Distribution

montague_nlu-0.1.6-py3-none-any.whl (19.2 kB view details)

Uploaded Python 3

File details

Details for the file montague-nlu-0.1.6.tar.gz.

File metadata

  • Download URL: montague-nlu-0.1.6.tar.gz
  • Upload date:
  • Size: 12.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.12.1 pkginfo/1.4.2 requests/2.18.4 setuptools/39.0.1 requests-toolbelt/0.8.0 tqdm/4.28.1 CPython/3.6.7

File hashes

Hashes for montague-nlu-0.1.6.tar.gz
Algorithm Hash digest
SHA256 f48741e23296f0e0a714d46b6c40e2dfc55e3af46bea103bff97af5bfa88c938
MD5 16cae95b86d1357ffaf71cfad1bd6408
BLAKE2b-256 78f576138a160ad3ab5c53ad7f68ab5b127437095b9620a3ed0baddae2336e50

See more details on using hashes here.

Provenance

File details

Details for the file montague_nlu-0.1.6-py3-none-any.whl.

File metadata

  • Download URL: montague_nlu-0.1.6-py3-none-any.whl
  • Upload date:
  • Size: 19.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.12.1 pkginfo/1.4.2 requests/2.18.4 setuptools/39.0.1 requests-toolbelt/0.8.0 tqdm/4.28.1 CPython/3.6.7

File hashes

Hashes for montague_nlu-0.1.6-py3-none-any.whl
Algorithm Hash digest
SHA256 7f6f3c9a2dce05d1e2bcb2099b73d5303bc1689edd609cb976d9e0f9f9916565
MD5 f2555a541fd6c362e8199a7bd4e2470a
BLAKE2b-256 44770ff048fc0e29ee7b0371520156ed1b237aeb967c852fb44ce3f63df2c026

See more details on using hashes here.

Provenance

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