Natural Language Processing API | Developer Portal | Cortical.io
Natural Language Processing API | Developer Portal | Cortical.io
The Cortical.io API Developer Portal gives subscribed Customers access to our natural language processing API powered by state-of-the art AI tools and Cortical.io‘s proprietary Semantic Folding method. Customers can choose between free or paid subscriptions and utilize all features included in the selected subscription(s).
Cortical.io is committed to providing reliable services and outstanding support to all Customers.
Cortical.io shall provide an annual average of 99% uptime availability for paid Products. This excludes any downtime for planned maintenance and interruptions beyond our control. Availability is calculated as the number of hours the services of the Products are functional, divided by the total number of hours within the respective calendar year, i.e., availability= (actual uptime / scheduled uptime)* 100.
Unavoidable downtime due to planned maintenance work will be communicated to the customer electronically (e.g. by email) in due time in advance.
Response times for API requests will be under 1000ms for 95% of the requests.
Customers can make up to 15 requests per second based on the chosen plan. Additional rates can be negotiated to accommodate the Customer's needs.
Scheduled maintenance will occur outside of peak business hours, typically between 2:00 AM and 4:00 AM Central European Time.
Cortical.io will do its best, within reason, to respond to defect or service interruption support requests within 24 hours of such notification and to solve the problem as quickly as possible.
The API uses Authentication Token for secure authentication and access to the API. You can find your key in your account settings. It is important to keep your key confidential. If your authentication key becomes compromised, you can create a new key and remove the old key in your account settings.
All data transferred between the Customer and the Cortical.io infrastructure is encrypted using state-of-the-art TLS encryption (TSL 1.3).
No Customer content is logged or permanently stored beyond the extent technically required for the operation of the API. Customer content or processed content is not used to train our models.
For the avoidance of doubt, Cortical.io shall be entitled to create and retain data logs for billing, security and statistical purposes. Such access logs shall not contain content or processed content. These logs may, however, contain metadata of API requests, such as the time of the API request and the size of the transferred content.
Cortical.io is also entitled to store and process the data automatically generated by the Cortical.io Software during the processing of the content uploaded by the customer. This is done in order to improve and customize the processing output in relation to the respective customer. After termination of the Subscription Contract, this data will be deleted immediately (a return to the customer is not possible due to its nature).
Cortical.io performs continuous monitoring for malicious activities, with alerts set up for any unusual patterns.
Data sent to the API is not stored or retained except for the minimum that is necessary for the operation of the API and legal purposes.
Data is backed up daily.
In case of a data loss event, recovery procedures will be initiated within 1 business day.
The API endpoints accept JSON-encoded request bodies and return JSON-encoded responses.
Detailed documentation for each API endpoint, including expected input parameters and return values, is available on the Development Portal by logging in with your registered user account and navigating to: https://api.cortical.io/portal/catalogue-products/nlp-api-product#/
Cortical.io applies semantic versioning for our API. Changes in the major version number indicate a backward-incompatible change. Changes in the minor version number indicate a backward-compatible change. Changes in the patch version number indicate a bug fix or other notable change.
Older versions of the API will continue to be supported for three months after the release of a new major version.
Customers will be notified of any upcoming changes or deprecations at least 1 month in advance.
Custom features or data fields can be added on a case-by-case basis, subject to feasibility and associated costs. Please contact our support team (api-support@cortical.io) to communicate feature requests or enhancements to existing functionality.
The Cortical.io Developer Portal offers dedicated portals for Customers to monitor their usage, manage API keys, and access detailed analytics.