IMPLEMENTING CHATBOTS USING DIALOGFLOW
DIALOGFLOW
Technology framework to build products/applications with ability of natural language conversations based human–computer interactions. Thereby, enabling users to interact with product by building engaging voice and text-based conversational interfaces, such as voice apps and chatbots, powered by AI.
Through Dialogflow, we can provide natural and rich conversational experiences to users on website, mobile app, the Google Assistant, Amazon Alexa, Facebook Messenger, and other popular platforms and devices.
Courtesy: Dialogflow, Google
REACH OF DIALOGFLOW
Build on any platforms- Build Actions, Skills, bots, and apps for the Google Assistant, Alexa, Cortana, Facebook Messenger and other platforms where users are on.
Build applications that work across devices- Whether users are on-the-go or at home, engage with them through wearables, phones, cars, speakers and other smart devices.
Global Reach — Connect globally with 20+ supported languages including Spanish, French & Japanese.
TECH/ARCHITECTURE CAPABILITIES
Powered by Google’s Machine Learning — Dialogflow incorporates Google’s machine learning expertise and products such as Google Cloud Speech-to-Text.
Built on Google Infrastructure — Dialogflow is backed by Google and runs on Google Cloud Platform, providing the ability to scale to hundreds of millions of users.
Optimised for the Google Assistant- Widely used tool to build Actions for more than 400M+ Google Assistant devices.
CHATBOT MODULES
Courtesy: Dialogflow, Google
NLP MODULE
NLP to enable classification of the intent of the chat conversation and route the flow to appropriate dialogue handlers.
BOT MODULES
Modules responsible for Dialog Management and Session Management of the conversation flow.
REPOSITORY MODULE
We will be implementing the proposed Chatbot on a retrieval based model wherein the solution will be using a repository of pre-defined responses and a heuristic implementation to choose the most appropriate response based on the input and context.
CHAT INTERFACE
Implement the chat interface on the web portal using Dialogflow.
CACHE
To store and manage the state that helps recognise the context of the input.
Dialog Management
Dialog Handlers to manage the conversation flow based on intent recognised by the NLP modules.
SESSION MANAGEMENT
Manage session context and data for a particular conversation flow.
