This is a minimal Python 3.x client library for accessing the Annif REST API which can be used for automated subject indexing and classification of text documents.
The easiest way to install is via pip:
pip3 install annif-client
The library depends on the requests module which is used for HTTP/REST access. If you install this via pip, the dependencies will be handled automatically.
The client library comes with examples demonstrating its usage. You can invoke the example by running the annif_client.py script.
In your own code, you can use the AnnifClient class like this:
from annif_client import AnnifClient
# then you can create your own client
annif = AnnifClient()
Here is the output from a typical example session:
$ python3 annif_client.py
Demonstrating usage of AnnifClient
* Creating an AnnifClient object
Now we have an AnnifClient object: AnnifClient(api_base='http://api.annif.org/v1/')
* Finding the available projects
Project id: yso-fi lang: fi name: YSO ensemble Finnish
Project id: yso-sv lang: sv name: YSO ensemble Swedish
Project id: yso-en lang: en name: YSO ensemble English
Project id: tfidf-fi lang: fi name: TF-IDF Finnish
Project id: tfidf-sv lang: sv name: TF-IDF Swedish
Project id: tfidf-en lang: en name: TF-IDF English
Project id: fasttext-fi lang: fi name: fastText Finnish
Project id: fasttext-sv lang: sv name: fastText Swedish
Project id: fasttext-en lang: en name: fastText English
Project id: maui-fi lang: fi name: Maui Finnish
Project id: maui-sv lang: sv name: Maui Swedish
Project id: maui-en lang: en name: Maui English
Project id: annif-api-fi lang: fi name: Annif prototype API Finnish
Project id: annif-api-sv lang: sv name: Annif prototype API Swedish
Project id: annif-api-en lang: en name: Annif prototype API English
Project id: ykl-fasttext-fi lang: fi name: YKL fastText Finnish
* Looking up information about a specific project
Project id: yso-en lang: en name: YSO ensemble English
* Analyzing a short text from a string
<http://www.yso.fi/onto/yso/p2228> 0.2595 red fox
<http://www.yso.fi/onto/yso/p5319> 0.2039 dog
<http://www.yso.fi/onto/yso/p8122> 0.1946 laziness
<http://www.yso.fi/onto/yso/p25726> 0.1285 brown
<http://www.yso.fi/onto/yso/p4760> 0.1220 triple jump
<http://www.yso.fi/onto/yso/p4758> 0.1194 long jump
<http://www.yso.fi/onto/yso/p2229> 0.1109 canines
<http://www.yso.fi/onto/yso/p10636> 0.1094 blue fox
<http://www.yso.fi/onto/yso/p4759> 0.1068 high jump
<http://www.yso.fi/onto/yso/p28336> 0.0911 animal training
* Analyzing a longer text from a file, with a limit on number of results
<http://www.yso.fi/onto/yso/p16495> 0.3651 licences (permits)
<http://www.yso.fi/onto/yso/p2346> 0.1656 copyright
<http://www.yso.fi/onto/yso/p11657> 0.1566 national libraries
<http://www.yso.fi/onto/yso/p6068> 0.1461 Apache
<http://www.yso.fi/onto/yso/p14833> 0.1220 copies
Try the demo at https://m.text-analytics.buildvoc.co.uk/
Instructions for using the app
-
First, take a picture of a document (or a part of it such as the introduction section) using your mobile device. Then the app will automatically:
-
Convert it to text using Optical Character Recognition (OCR) technology.
-
Analyse the text using Annif Present you with a list of suggested subjects of interest that have been detected in the article. Also will provide you with search results for fire safety guidance.
Take photo of text content with a webcam (code snippet from Colab), run optical character recognition on it with Tesseract (wrapper for Google’s Tesseract-OCR), and send the text to Annif.
The code is published under the Apache 2.0 license.