Platforms for Chatbots in a Post-App World
Platforms for Chatbots in a Post-App World
As technology matures, options only grow. It's possible that before long, a majority of user interactions with your apps will be conducted via chatbot interfaces.
Join the DZone community and get the full member experience.Join For Free
As the Internet moves towards a post-app world, a world where content is delivered via bots, many different types of content providers are rising to prominence. The most popular of these providers will be the basis for the continued growth of the post-app Internet, and as such it makes sense to get a feel for what capabilities are currently available for bot-writers looking to leverage one of the more popular sites to drive their efforts. Below, we’ll look at a few of the options available for integrating with a bot, and deep-dive into one — Facebook Messenger — to see what an integration might look like.
Fertile Ground for Bots
The core interface for the post-app world will be textual – whether that text is typed in from a keyboard, or spoken aloud and processed by an intermediary. This means that social media providers will offer some of the first examples of what intelligent bots can achieve. Slack, a popular chat program used by many businesses for internal and external communication, offers an API for developers to write their own intelligent plugins for the chat platform, offering these apps via a dedicated portal for plugins. Telegram is another communication medium that provides an API that bot writers can use to automate message processing, present customized advertising, or perform online purchases on behalf of the user. In the remainder of this article, we’ll be digging into Facebook, the most popular social network. Facebook offers a powerful API that integrates with Messenger, allowing you to quickly get up and running with a Facebook messaging bot that can enhance users’ experiences both on Facebook itself and in your application.
Deeper Dive Into Chatbots on Facebook Messenger
Facebook’s Messenger API offers developers the capability to integrate an automated chat bot with the Messenger chat window, giving developers the opportunity to respond to messages in an automated fashion. Once a user has granted you bot access, you can immediately begin communicating with them via Messenger using only their phone number. Facebook provides a robust event-driven API that allows you to receive and respond to messages in real-time, communicating with your application via a series of web hooks and presenting the results to the chat user. Their API documentation offers a comprehensive quick-start guide that quickly builds an automated echo bot, communicating with your application via standard HTTP messages. The use of structured messaging also allows you to integrate app elements into the chat experience, using tools like buttons and hyperlinks to drive an event-driven application. Facebook has released a demo application that shows a lot of the functionality that Facebook has to offer for bot writers.
Once you have a Messenger bot up and running, you are free to add as much functionality as you like. One common option is to integrate with Wit.ai, a service that provides deep-learning text analysis. Using Wit.ai, you can easily add natural language processing to your bot. This allows you to dynamically suggest products based upon user messages, perform automated actions based on the perceived user intent, or any of a number of other types of functions. Their Getting Started docs provide details of the functionality available, but the true power lies within their HTTP API. By modifying your bot’s backing app to communicate with Wit.ai via HTTP calls, you can easily add intelligent automated language processing to your bot, providing a more feature-rich and engaging experience for users interacting with you.
In the post-app world, the vast majority of communication and user interaction is likely to happen through one of a handful of communication providers. These can be chat programs like Slack, or full social media experiences like Facebook – really any program that offers a natural language-based mode of interaction.
Most of the major social media providers offer APIs with which you can integrate, making your products, advertisements, and anything else you may wish to offer available to the user via a set of callbacks and responses. Above, we looked into what Facebook has to offer, but each of the major social media providers offers an integration channel for a custom application, allowing you to leverage internet-scale tools like Wit.ai for language processing and intent-based communications.
As the technology matures, options and offerings will only grow, and it is possible that before long, a majority of user interactions with your applications will be conducted via these chatbot interfaces.
Published at DZone with permission of Itay Herskovits , DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.