
<h2>API operations</h2>
<p>
All API operations can perform analysis on supplied raw text or on text extracted from a given URL.
Optionally, one can supply additional information which can make the result more precise. An example
of such information would be the language of text or a particular text extractor for URL resources.
</p>
<p>The supported types of analyses are:</p>
<ul>
<li><strong>lemmatization</strong> ⟶
Finds out lemmata (basic forms) of all the words in the document.
</li>
<li><strong>correction</strong> ⟶
Performs correction (diacritization) on all the words in the document.
</li>
<li><strong>topic detection</strong> ⟶
Determines a topic of the document, e.g. finance or sports.
</li>
<li><strong>sentiment analysis</strong> ⟶
Determines a sentiment of the document, i.e. how positive or negative the document is.
</li>
<li><strong>named entity recognition</strong> ⟶
Finds named entities (like person, location, date etc.) mentioned the the document.
</li>
</ul>
<h2>Encoding</h2>
<p>The supplied text is expected to be in UTF-8 encoding, this is especially important for non-english texts.</p>
<h2>Returned values</h2>
<p>The API calls always return objects in serialized JSON format in UTF-8 encoding.</p>
<p>
If any error occurs, the HTTP response code will be in the range <code>4xx</code> (client-side error) or
<code>5xx</code> (server-side error). In this situation, the body of the response will contain information
about the error in JSON format, with <code>exception</code> and <code>message</code> values.
</p>
<h2>URL limitations</h2>
<p>
All the requests are semantically <code>GET</code>. However, for longer texts, you may run into issues
with URL length limit. Therefore, it's possible to always issue a <code>POST</code> request with all
the parameters encoded as a JSON in the request body.
</p>
<p>Example:</p>
<pre><code>
POST /s1/sentiment
Content-Type: application/json
{"text":"There is no harm in being sometimes wrong - especially if one is promptly found out."}
</code></pre>
<p>This is equivalent to <code>GET /s1/sentiment?text=There%20is%20no%20harm...</code></p>
<h2>Request limitations</h2>
<p>
The API has other limitations concerning the size of the HTTP requests. The maximum allowed size of any
POST request body is <em>512 KiB</em>. For request with a URL resource, the maximum allowed number of
extracted characters from each such resource is <em>100,000</em>.
</p>
<h2>Terms of Service</h2>
<p>
By using the API, you agree to our
<a href="https://www.geneea.com/terms.html" target="_blank">Terms of Service Agreement</a>.
</p>
<h2>More information</h2>
<p>
<a href="https://help.geneea.com/index.html" target="_blank">
The Interpretor Public Documentation
</a>
</p>
No direct contact info provided by the API owner.
Looking to integrate Geneea Natural Language Processing into your project? Our team can help you build custom connectors.
Get Integration Help