Saplo provides services that extract and refine valuable information from large quantities of text. We offer a range of cost efficient text analysis services through an API (Application programming interface). Services can be used either stand alone or combined into new and exiting concepts and technologies.

Documentation and guides for Saplo API is found on the developer page: http://developer.saplo.com.

Getting Started

Install

pip install git+git://github.com/rogeliorv/pysaplo.git

Or download https://github.com/rogeliorv/pysaplo/zipball/master and run

python setup.py install

Get your API key http://saplo.com/api

Import

from pysaplo.saploapi import *

Authentication

To authenticate with the API you run the method auth.accessToken. If your API key and your Secret key (your key pair)

is valid a token is created which you will use for further API calls.

When creating the SaploJSONClient object you will automatically be authenticated.

client = SaploJSONClient("YOUR API KEY", "YOUR SECRET KEY")

You can also explicitly get your Access Token

print client.token

To be able to extract information (Tags, Related Texts, Related Contexts etc.) from the API you first need to add some

texts to the system. Texts are stored in text collections that you create.

A collection can only contain text for a single language (e.g. only English texts).

collection_id = client.collection.create( name = "My Collection", description = "This is my first collection for english texts.", language = "en" )

print "New collection created with Id:", collection_id

Add a text to a collection

After creating a collection you can start adding texts to it.

Add a text using the method text.add. When the text has been added it will get a unique id inside of that collection.

text_id = client.text.create( collection_id=678, headline='Example Headline', body='This is a short example text. Insert a longer one for better results' )

print "New text created with Id:", text_id

Get Entity Tags

Now that we have a text in our collection we can get entity tags from that text.

Get tags using the method text.tags

tags = client.text.tags( collection_id=678, text_id=2, wait=10 )

print "Retrieved these tags from the text:", tags

Get Related Texts

To be able to get related texts we need to add additional texts to the collection. When fetching related texts the API

searches through the collection to find the texts that are most similar to each other.

To get related texts we run the method text.relatedTexts and specify the text id and collection id we want results for.

The system will now index all texts and give return a list with results. In the result list you will find the related

text id and the relevance which is a value between 0.0 and 1.0, a higher number indicating more relevance.

related_texts = client.text.related_texts( collection_id=678, text_id=1, wait=30 )

print "Related texts:", related_texts

Reset a Collection

reset_collection = client.collection.reset( collection_id=678, )

print "Successfully reseted corpus with properties: ", reset_collection