What are Conversational Interfaces (CI) and their Benefits

Although no chatbot has passed the Turing test yet (or so we assume…), it’s clear that the technology has developed rapidly in recent years. Chatbots like SmarterChild, the artificial “buddy” users could communicate with via AOL Instant Messenger, bear little resemblance to sophisticated products like Siri and Alexa.

Companies throughout numerous industries have triggered this rapid growth by funding the development of chatbot technology designed to make interactions more closely resemble conversations with an actual human. Until recently, Natural Language Processing (NLP) technology supported most of these products. NLP innovations allowed chatbots to comprehend and respond to basic user requests.

That being said, the results companies get from the NLP approach have been far from perfect. No one is going to mistake an NLP-based chatbot for a real person. Thus, organizations have shifted towards a Natural Language Understanding (NLU) approach, hoping to develop chatbots that can learn how to understand language more efficiently and reliably.

Still, the results can be cumbersome. Chatting with a website’s bot is often a slow, drawn-out process. In the age of digital distraction, chatbots should be able to communicate with users through a variety of means.

That’s why many experts are beginning to recommend Conversational Interfaces (CI) as the ideal solution.

Understanding CI: The Basics

Essentially, a CI is a hybrid chatbot interface. It incorporates voice, text, graphics, and any other available language interface to facilitate the most efficient communication between user and chatbot. By using multiple forms of communication, this type of chatbot also guards against user distraction.

Slack’s chatbot is a good example of a CI. One of the distinguishing characteristics of a CI is compatibility with multiple devices and platforms. You can interact with a CI through your phone, tablet, laptop, and even smartwatch. The ideal CI chatbot will respond to both voice and text, whichever is more convenient for you at the time.

A CI-based chatbot also reduces the frustration a user experiences when they’re unable to explain to a bot what it is they’re trying to do. For example, many people have encountered chatbots on websites that introduce themselves with general inquiries, like “How can I help you today?”

The problem is, the chatbot often doesn’t understand the user request. The syntax of your request must be comprehensible to a computer.

With a CI chatbot, instead of asking a general question, the bot may offer users a list of options. While this might seem like it limits what a person can actually use a chatbot for, it actually makes the bot more functional in a practical sense; the less chances there are for confusion, the less chances there are for a user to get frustrated with the service.

To understand the potential applications of this type of chatbot, take a look at Typeform. It’s an informational article that allows users (if they choose) to occasionally stop reading in order to interact with a chatbot that provides information about the topic. In the future, companies could use this kind of chatbot to gather more personal information about a customer. Imagine reading about a new product or service, while a reasonably natural chatbot asks you questions, answers your questions, and generally takes the time to learn about your reaction to the product.

It’s important to realize that CI technology is still in its infancy. That said, it represents one of the most promising innovations for developing truly realistic virtual people.

Post a Comment