Oracle Digital Assistant

Artificial Intelligence – Built In AI Algorithms with Machine Learning, Context, Conversation Flow

Channel Configurator

Channel Configurator End users have a choice of messaging channels they prefer to use. Certain geographies have specific channels as their preferred messaging channel while other geographies prefer other messaging channels. There are different categories of channels as well. Broadly, these can be categorized as follows: • OTT Channels: Over the top (OTT) messaging channels such as Facebook Messenger, Facebook WhatsApp, WeChat, Line, Kik, Telegram, Talk, Skype, Slack, SMS. • VPA: Virtual Private Assistants such as Amazon (Dot, Echo, Show), Google Home, Apple HomePod. • Mobile & Web app extensions: Extending native or hybrid/responsive mobile apps or web applications with chat capabilities. • Voice Based Input: Custom devices or apps with interfaces that use Siri, Cortana, Google Voice or other speech input for interaction.

Dialog Flow Execution

A conversation with a chatbot goes through a specific flow with the conversation having different states and context. This flow defines what should happen next based on an input and the flow itself is implemented in the platform as a state machine, which can be thought of as a workflow or process flow. The steps are the states in the workflow which the system guides the end user through as part of the conversation. The system that guides you through those states is called a state machine. To define your flow in the chatbot, you will define specific states and assign actions (called Components) to do work in that state. These steps or states take different paths based on the user input that impacts the decision the chatbot makes for the flow. Intents and Entities are the user-configurable components of the machine learning (ML) based natural language processing (NLP) algorithm that is used in the chatbot flow to manage the state or the steps. Variables store values that can change, depending on conditions or on information passed to the system. The Oracle Intelligent Bots Process Flow in chatbots is a ‘dialog flow’, as you will be creating applications that interact with users in the form of a dialog. The process itself is implemented as a state machine via a set of rules and it is defined in the Intelligent Bots Platform console in a format known as Oracle Intelligent Bots Markup Language, Intelligent BotsML for short. Intelligent BotsML is a form of YAML. YAML is a markup language that is, fundamentally, a way of expressing data and data relationships in a format that is highly focused on human readability. One of the differentiated value of this dialog designer and execution engine is that the dialog designer and execution engine is tightly integrated with the ML based NLP engine. Human conversations are often non-linear in nature. End users could potentially branch into different states / context in course of a conversation. For example, let’s say I want to transfer funds from account A to someone. Let’s say I start by asking the chatbot – pay Tom for dinner. The chatbot responds with “from which account?”. Let’s say the user picks checking but realizes he is not sure of how much he has in the account; he switches context to ask for balance and further wonders about and asks for recent transactions, and so on. In other words, the user triggers changes in the state, from transferring money, to checking balance, and then to recent transactions. At some point, he decided to pay Tom, which is the original state he started with. The Oracle Intelligent Bots platform makes it very easy to model this with built in state management so the developer does not have code and maintain the complexities of state management, reducing the time to deploy and the cost of ownership in managing the code.

Artificial Intelligence Engine

The Oracle Intelligent Bots expose multiple natural language understanding (NLU) models to predict user-intent from incoming chatbot requests and accurately execute the required dialog flow. To achieve this, we utilize multiple natural language processing (NLP) and machine learning (ML) algorithms combined with other approaches to classify end user intent.

Custom Components

Custom Components is another name for ‘3rd party integrations’. Custom Components can consume any integration service (REST) for almost any ERP, CRM, E-Commerce, Warehouse Management Software, Transportation Management Software or Bot-to-Live Agent hand off to any ACD platform. For chatbots to add value to a conversation, it must be able to integrate to the system of record that holds information required for the targeted use cases. These are typically enterprise backend systems that own customer’s banking information, or employee information in a HCM system, or sales information in a Sales CRM system. Oracle is the leader in integration to enterprise and cloud data sources and this is also provided out of the box, reducing the effort, time and cost to expose data to the chatbot. With built in integration to Oracle Mobile Cloud, Enterprise (OMCe), the Intelligent Bots Platform can leverage the shaped mobile APIs, and in addition, making use of mobile-first services, like Push Notifications to send asynchronous notifications from the backend systems to the chatbot.

Intelligent Bot Analytics