Oracle Digital Assistant is an advance chatbot program that is designed to simulate complex conversations with human users. It uses artificial intelligence (AI), natural language processing (NLP) and machine learning (ML) to learn. It helps to improve efficiency, provide better assistance to their users, either employees or customers, and reduce operational costs as well. There are many features that help in automation, increase productivity, and improve user experience for various Oracle SaaS applications such as Human Capital Management (HCM), Supply Chain Management (SCM), Finance, Customer Experience (CX) etc. It can create huge automation opportunities in IT service operations to optimize overall support cost and improve year-on-year efficiency.

Oracle Digital Assistant is a part of Oracle Cloud Infrastructure (OCI) PaaS (Platform as a Service) , which is an AI-powered, highly secure, and cost-effective service. It can use the Oracle Identity cloud service (IDCS) that depends on the service license. It supports VBCS for single page application. It has multilingual capabilities; skills can develop either in single-language or multi-language.

• Pre-Requisite

• Integration

  There are two integrations required to communicate with the user and perform transactions.

  • Channel Integration

    Channel Integration is a configuration that creates communication interface between user and Oracle Digital Assistant for conversation. There are different types of channels such User, Agent Integration, Application and System.

    Multiple User Channel types such as Facebook Messenger, MS Team, Web etc., are available that can be used to communicate. Microsoft Team and Facebook Messenger have Oracle Digital Assistant native support which can easily integrated with Oracle Digital Assistant.

  • Backend Integration

    Backend integration is used to retrieve and update data from Oracle Digital Assistant to any external service that could be cloud application, on-premises application which available on internet.

    Oracle Digital Assistant skills can be integrated with custom component that need to develop using Node.js and Java Script. Custom component calls the REST APIs (for example, Oracle HCM Cloud REST APIs) to get and send data to HCM Cloud Application.

    This will be deployed as package as an embedded component service in Oracle Digital Assistant.

    REST APIs can be authenticated either through Bearer token or Basis authentication i.e. Cloud application credentials, Bearer token authentication recommended by Oracle, for this, IDCS required.

• Security

  Oracle Digital Assistant mainly deals with enterprise data such as personal data, financial data and sales data which is very sensitive for any organization, so it is using multi-layer security mechanism to protect data.

  • Channel Security

    The security between the user device and the message server is the responsibility of the message channel provider, for example, Facebook or MS team.

  • Oracle Cloud Infrastructure (OCI) Security

    OCI Identify and access management (IAM) manage user (platform user like admins, developers etc.), resource group (compartments), and policies.

  • Application Security

    Oracle Digital Assistant uses OAuth2 authorization for application security either using authorization code (Token) or Cloud application credentials. There is setup required to configure an authentication service to enable interaction between Digital Assistant and the identity provider such as with IDCS (Oracle identity cloud Service). Oracle SaaS REST and SOAP APIs are secured with role-based security that can access using Cloud application credentials and bearer token-based authentication.

• Flow Design

  • User Interface (UI) Design

    Oracle Digital Assistant user interface (UI) should be friendly and easy to use. All services should be menu based that listed different services, the user can select that service to start the conversion to get the service. User-friendly UI will help to communicate easily and avoid any misunderstanding.

    The routing model determines how Oracle Digital Assistant routes the conversion based on user input.

  • Intent Design

    Intents allow your skill to understand what the user wants it to do. An intent categorizes typical user requests by the tasks and actions that your skill performs.

    Intents are comprised of permutations of typical user requests and statements, which are also referred to as utterances. A rich set of utterances enables a skill to understand what the user wants when it receives ambiguous user input.

    Oracle Digital Assistant can mature if increase the number of utterances. We need to add utterance as much as possible, at least 20-40 high quality.

  There are some systems pre-defined intent such Help, Exit and UnresolvedIntent.

  • Dialog Flow Design

    Dialog flow is a model of the interaction between the user and Oracle Digital Assistant, this can be designed either using visual flow designer or OBotM, a special implementation of the YAML.

    OBotML is old method to design YAML.

    With the Visual flow designer, dialog flows are designed visually instead of through code editor. The connections between states are represented in visual diagrams. Variables, parameters, and component properties are defined in specialized editors and dialogs.

• Pre-built Skills

  Oracle Digital Assistant is available with ready-to-use pre-configured skills for various services that are already integrated with the SaaS applications. These pre-built skills can be extended.

  There will be separate blog on Oracle Digital Assistant Cloud service configuration, Channel integrations and Custom Component development (Calling REST APIs to perform transaction on HCM cloud application).

  References

  https://docs.oracle.com/en/cloud/paas/digital-assistant/

Dialog Flow Design using OBotML Code in Flow editor

Demonstration

Custom Oracle Digital Assistant Integrated with Facebook Messenger channel