Hydra Scanners are A.I. models that you can create using a simple UI. They connect to a data source (e.g. Zendesk, Box, Segment, etc.) to source data and make predictions to provide intelligent automation capabilities in various business workflows. You can train Hydra Scanners to fit your unique business needs. Below are a few examples of how you can use Hydra Scanners:
i. Extract metadata from documents
ii. Classify documents
iii. Predict customer churn
iv. Predict high-value customers
v. Predict customer personas
vi. Flag time-sensitive tickets
The process of creating a Scanner includes the following steps:
i. Identifying the appropriate Scanner Template
Hydra provides a series of pre-created Scanner Templates covering a wide range of use cases. As an example, there is a Scanner Template for Extracting metadata from documents in the Box platform. If you don't see a Scanner Template that fits your needs, please ask. We are happy to help.
ii. Providing a name and a description
The name and description help you identify your Scanner on the Scanners page. We recommend using short and descriptive names and descriptions.
iii. Creating a label set
A label is a name that you use to recognize a data point that fits a specific pattern. As an example, if you are building a Scanner to extract metadata from invoices, your labels are likely to include: invoice number, PO number, invoice date, etc. If you are building a Scanner to categorize customers by their revenue potential, the labels are likely to include a ranking such as Tier 01, Tier 02, Tier 03, etc.
iv. Configuring data filters
Data filters provide the option to configure how you source data from the source system. The configuration options depend on the Scanner Template you are using.
v. Configuring actions
Actions provide the options to execute a task using the Scanner's output (i.e. predictions). The available actions list is dynamically populated based on the Scanner Template in use and the Connections available on your workspace. As an example, you can create a Salesforce opportunity using the data extracted from a document using a Hydra Action.
vi. Training and activating the Scanner
Hydra Scanners are user trainable. Hydra shows the appropriate trainer UI depending on the Scanner Template you selected during the Scanner creation. The available trainer UIs are document trainer, document category trainer and data column trainer.