Frequently Asked Questions
Support Intelligence

What types of information does the Cortical.io Support Intelligence Engine search and process?

Support cases, service notes, user forum posts, technical manuals, training material, known issues, and other related knowledge sources

In which languages does the engine search and process information?

English. The functionality to search and process other languages can be added on request.

Is learning by the engine supervised or unsupervised?

The Support Intelligence Engine learns the vocabulary of your company’s business domain by analyzing a corpus of relevant information sources—for example, existing support cases and product documentation—in an unsupervised machine-learning approach. The engine can also learn by analyzing support agent ratings of results according to relevancy. The system continuously analyses the ratings and learns which support case fields to give most weight to when ranking top results for subsequent searches.

How does the engine handle non-relevant text in support cases?

All support cases are automatically filtered so that non-relevant text, such as generic introductions, greetings, and email signatures, and duplicate text are removed. The Support Intelligence Engine identifies variations of this type of text throughout all support cases, without requiring exact text searches for each variation.

How are the search results visualized?

Structured output can be directly integrated into your company's existing systems through the engine's REST API. The engine itself does not include a user interface: the results of the data extraction feed into your support management workflow, affecting information that your support agents view in their user interfaces.

Is the engine easily retrained?

As new product names, feature names, and other technical terms enter the jargon of your agents and customers, the Support Intelligence Engine can be easily retrained (for example, every 6 or 12 months) in the updated domain vocabulary of your business. Retraining can be done in parallel with normal functioning of the engine and takes only a few hours.

How does the engine integrate into my existing support case software?

Through the engine's REST API.

How is the engine deployed?

The engine can be installed on a server on your company’s premises or in your company’s private cloud. Cloud production environments are currently operating on Google Compute Engine (GCE) and Amazon Web Services (AWS) instances.

Still have some questions? Contact us to get the answers!