New release of MeaningCloud APIs

We keep improving the functionality and user experience of MeaningCloud, our horizontal APIs, to offer you a more powerful and easier to integrate text analytics technology.

In this new release, besides incorporating many of the suggestions you have been sending us through our online support, we included several improvements:

  • We extended the possibility of using custom dictionaries to the Sentiment Analysis and the Spell, Grammar and Style Proofreading APIs
  • We optimized and uniformized the user experience and the documentation across the various APIs. 

Improvements to our horizontal text analytics API

These are the changes affecting all APIs:

  • The output has been homogenized to follow the new standard for array elements in all APIs (an array of objects will be called “[object]_list”).
  • Error messages in all APIs have been unified.
  • The response headers have been updated so that the content type reflects correctly all output formats supported.
  • Several minor bugs have been fixed, and linguistic resources have been updated.
  • The documentation has been improved, both in format, content and page response time.

And these are the additional changes that affect some of the APIs only:

  • Topics Extraction API; Lemmatization, POS and Parsing API:
    • Related Facebook and Twitter links have been added to the semantic linked data information (semld) of known entities and concepts.
    • Heuristic rules for entity detection have been improved, increasing the quantity and the classification quality of the unknown entities detected.
  • Sentiment Analysis API:
    • Irony detection has been added both at a global level as well as at segment level.
    • Entity/concept ID and type added to the output.
    • User dictionary parameter, ud, added.
  • Spell, Grammar and Style Proofreading API:
    • Several changes in the parameters:
      • Four more specific new parameters (psr, nsr, bsr and str) have been added in Spanish to deal better and separately with the different cases of consonant reduction.
      • A new parameter have been added, ilang, that allows the user to define the language in which the messages in the response appear.
      • Two new parameters related to non-mandatory accent marks in Spanish have been added (sdm, ddm).
      • A new parameter related to vowel reduction in Spanish has been added (vr).
      • The “Warn of percentage signs not spaced” parameter, wps, has been limited just for Spanish.
    • The possibility of using a user dictionary, ud, has been added.
    • Messages associated to the different issues found in a text have been improved.
    • General improvement in grammar checking in English.

As we anticipated in our last newsletter, this update means that older releases will no longer be operational by July 1, 2014.

This is a list of the API versions to be deprecated and the versions that you should be using instead:

API Deprecated version Latest version
Topics Extraction 1.1 1.2
Text Classification 1.0 1.1
Language Identification 1.0 1.1
Sentiment Analysis 1.0 1.2
Spell, Grammar and Style Proofreading 1.0 1.1
Lemmatization, POS and Parsing 1.1 1.2

We hope you like this new release. And if you are not a MeaningCloud (formerly known as “Textalytics”) user don’t forget to register and process for free up to 500 000 credits per month.

Thank you for continued support!

Leave a Reply

Your email address will not be published. Required fields are marked *