Unicorn NLP

Deep AI for Travel Reviews

Unicorn NLP

Deep AI for Travel Reviews

Human-Like Sentiment Analysis for Hotel Reviews API
(AI In-the-Box for Hotel Reviews)

UnicornNLP API converts any unstructured Review (or any other text) into structured ready-to-use data which could be easily quered and analyzed and displayed - in the same way like normal structured data. It could be used for enhancing user interface, analytics, next-gen reviews presentation, summary, additional filters for browsing or semantic search based on reviews and more.

The Human-like Analysis for Hotel Reviews API (accessible via RapidAPI) is available for Hotel Reviews. For other reviews, we are selling the On-Premise version with 0 cents / text.

Link to API on RapidAPI

Quick Description for Human-like Analysis for Hotel Reviews API

The API is simple to use. You put text (typically a review) that you want to process and you get a list of structured data detected in this text. Below, you will find the description of each field, both input and output. API is available via RapidAPI (link below) so you need to apply RapidAPI Key that you get after you sign up.


Input (POST):
If you have a standard title+text review, use only title and text field. If you have title, positives and negatives input field of a review, use title, positives and negatives and leave text empty. If you have title, text, positives and negatives you can use all the input fields. They are all optional, so you could leave some empty.


Output (JSON):


Example of a short review processed by our API:

Input (POST):

Output JSON - here displayed in a more pleasing for to human eyes. In this case, API detected 4 Recipes inside this text.

API OUTPUT JSON (after json_decode and converting to array)

array (
  'param' => array ('status' => 1),
  'output' =>    array (
    0 => 
    array (
      'id_recipe' => 61,
      'name_recipe' => 'bathroom_dirty',
      'id_counter' => 60,
      'category' => 'bathroom',
      'aspect' => 'bathroom',
      'feature' => 'dirty',
      'polarity' => -2,
      'segment' => 'Bathroom was a little bit dirty'
    ),
    1 => 
    array (
      'id_recipe' => 127,
      'name_recipe' => 'staff_friendly_helpful',
      'id_counter' => 129,
      'category' => 'service',
      'aspect' => 'staff',
      'feature' => 'helpful, friendly',
      'polarity' => 2,
      'segment' => 'but the service was very friendly'
    ),
    2 => 
    array (
      'id_recipe' => 213,
      'name_recipe' => 'tv_bad_old_poor_channels',
      'id_counter' => 245,
      'category' => 'addons',
      'aspect' => 'tv',
      'feature' => 'bad, poor channels',
      'polarity' => -1,
      'segment' => 'There was only 10 TV channels'
    ),
    3 => 
    array (
      'id_recipe' => 212,
      'name_recipe' => 'fitness_center_good_great',
      'id_counter' => 244,
      'category' => 'addons',
      'aspect' => 'fitness',
      'feature' => 'great',
      'polarity' => 1,
      'segment' => 'but the fitness was good equiped'
    ),
  ),
);

Go to Rapid API and you can start using this API today.

Link to API on RapidAPI

https://rapidapi.com/unicornNLP/api/human-like-sentiment-analysis-for-hotel-reviews