Automated text classification and answering platform

ARIS is an AI-powered platform, which understands any text message written in a natural language. The answers and all necessary actions are suggested automatically allowing the operators to instantly react on messages in a single click.

40 % - 50 %
of email replies can be completely automated
80 %
of messages are correctly categorized
75 %
ready-to-send answers

Advantages of ARIS in practice

  • Automatically categorizes 90% of messages.
  • Response and follow-up actions suggested automatically.
  • All triggered by one click.
  • Easy plug-in and deployment.

Businesses are facing an overwhelming amount of email communication. The emails on the same topic often repeat again and again and the customer support representatives spend their time replying.

ARIS platform automatically sends replies to repetitive emails. ARIS is not a chatbot. For complex emails, the platform prepares answers for the customer representative who just confirms it.

Based on previous deployments, we do see that about 40% - 50% of email replies can be completely automated.

How we classify automated text step by step

1) Adapt for business needs

We pre-process historical messages or other text data and eventually train our state-of-the-art NLP and Entity Extraction models. This way, the platform is automatically tailored to the clients' needs.

2) Visualization

All incoming messages, suggested responses and follow-up actions can be handled and managed through a single intelligent enhanced inbox. Customer service does not need to use multiple systems.

3) Integration

The platform can be integrated into a current environment setup and communicate with different third-party systems via API.

4) Continuous improvement

Based on the users' actions and feedback the platform continuously learns and improves its accuracy automatically.


Find inspiration for your business with our success stories



  • Text messages, emails, documents, news and articles automatic classification.
  • Find all key information in text.
  • Process significant amount of messages.
  • Minimal effort to learn any domain.
  • Multiple languages support.
  • Automated organizing and sorting.


  • Automatic reply construction in natural language.
  • Continuous improvement.
  • Instant replies on messages.
  • Reduce of load on customer support operators.
  • Increase efficiency of text communication.


  • Necessary actions suggestion based on the message content.
  • Perform all actions automatically.
  • One place to handle all changes.

  • Easy integration of current infrastructure.
  • One point to all operations.
  • Reduce costs on customer support.

Case study

How we correctly categorized more than 80 % of the incoming messages for P2P landing platform provider

Client: Peer-to-peer landing platform provider

Automatically process a large number of incoming customer support emails.

Based on historical data train a natural language processing model in order to classify and identify useful information in incoming emails. Define templates for automatic replies and actions executing.

More than 80 % of the incoming messages are correctly categorized from day one.

Area: Banking, fintech, natural language processing, customer support


Do not miss

If you want to categorize 90 % of messages automatically, contact us:

Thank you

We will contact you as soon as possible.

Štěpán Kopřiva

CTO Blindspot Solutions