Skip to content

datatxt.rst

Latest commit

 

History

History
101 lines (73 loc) · 3.59 KB

datatxt.rst

File metadata and controls

101 lines (73 loc) · 3.59 KB
title:The dataTXT API

The dataTXT API

dataTXT is a family of semantic services developed by SpazioDati. All its methods are available in the same class:

>>> from dandelion import DataTXT
>>> datatxt = DataTXT(app_id='', app_key='')

NEX: Named Entity Extraction

dataTXT-NEX is a named entity extraction & linking API that performs very well even on short texts, on which many other similar services do not. dataTXT-NEX currently works on Italian and English texts. With this API you will be able to automatically tag your texts, extracting Wikipedia entities and enriching your data.

You can extract annotated entities with:

>>> for annotation in datatxt.nex('Oh my, arduino is super cool, so #opensource').annotations:
...     print(annotation.uri)
http://en.wikipedia.org/wiki/Arduino
http://en.wikipedia.org/wiki/Open_source

Additional parameters can be specified simply by:

>>> result = datatxt.nex('Oh my, arduino is super cool, so #opensource',
...                      include_lod=True,
...                      )
>>> [annotation.lod.dbpedia for annotation in result.annotations]
['http://dbpedia.org/resource/Arduino',
 'http://dbpedia.org/resource/Open_source']

Check out the dataTXT-NEX documentation on dandelion.eu for more information about what can be done with NEX.

SIM: Text Similarity

dataTXT-SIM is a semantic sentence similarity API optimized on short sentences. With this API you will be able to compare two sentences and get a score of their semantic similarity. It works even if the two sentences don't have any word in common.

You can compute the semantic similarity between two texts with:

>>> datatxt.sim('Barack Obama is the president of the US',
...             'Bob Iger is the CEO of Walt Disney')
{'lang': 'en',
 'langConfidence': 1.0,
 'similarity': 0.2564,
 'time': 11,
 'timestamp': '2042-01-01T01:02:03'}

Check out the dataTXT-SIM documentation on dandelion.eu for more information about what can be done with SIM.

LI: Language Identification

dataTXT-LI is a simple language identification API; it is a tool that may be useful when dealing with texts, so we decided to open it to all our users. It currently supports more than 50 languages.

You can identify the language of a text with:

>>> datatxt.li('mamma mia! un testo in italiano!')
{'detectedLangs': [{'confidence': 0.9999952605110598, 'lang': 'it'}],
 'time': 0,
 'timestamp': '2042-01-01T01:02:03'}

Check out the dataTXT-LI documentation on dandelion.eu.

SENT: Sentiment Analysis

dataTXT-SENT is a sentiment analysis API that analyses a text and tells whether the expressed opinion is positive, negative, or neutral. Given a short sentence, it returns a label representing the identified sentiment, along with a numeric score ranging from strongly positive (1.0) to extremely negative (-1.0).

You can identify the sentiment of a text with:

>>> datatxt.sent('I really love your APIs')
{"sentiment": {
    "type": "positive",
    "score": 0.9
},
"lang": "en",
"time": 0,
"timestamp": "2018-10-11T14:45:15.529"}

Check out the `dataTXT-SENT documentation on dandelion.eu`_.