Chatbots - So hot right now!

These days, buzzwords such as “A.I.”, “Machine learning”, “Virtual Reality” and “chatbots”are talked about regularly on the internet as well as in the mainstream media. Chatbots are one of the more recent developments seen in web development and you can see it popping up all over the place in 2021 – For good…but also sometimes not not so good. For example a recent incident occurred with Microsofts’ chatbot “Tay”, who went on a racial/genocidal tweeting spree only 24 hours after it was launched.

Chatbots are all around us on the web, whether you are trying to reach customer support to change your mobile data plan, or when you need to fill in a questionnaire before being able to book an appointment at your local restaurant. These are just some of the applications that make use of chatbots over regular human interaction.

Chatbots can make it easier for users to find the information they need, but they can also assist businesses in their day-to-day work processes since chatbots can work 24/7.

What types are there?

There are many different types of chatbots, all using different technologies at their core. But in simplest terms there are two types: Basic and Advanced.

Basic

Basic chatbots are the ones where the user interacts with the program via predefined options. Based on the chosen option/answer the program knows what to do next. Chatbots that work according to this are also often called “Rule-based” chatbots, the program follows a “If This Then That” cycle to determine what response to give back to the User.

If the user clicks on answer A, then I will send back the store opening hours.

These types of chatbots are basic since they do not have many scaling options, their capabilities are very limited; though very useful and effective if it is what you need (and don’t get me wrong, since 70-80% of common enquiries are made up of very simple and short question-answers).

Some examples of a very basic chatbot are:

• Change user phone data plan by sending a SMS to a specific number
• Act as a interactive FAQ-supplier, so the user doesn’t need to search and click as much but can be given the response depending on their answer

Advanced

Advanced chatbots software is where we enter the sci-fi realm, where there are no limits (well… there are but more on that later). Natural Language Processing (NLP), Artificial Intelligence (AI) and Machine Learning (ML) are all terms that come into play now. These (and more) are advanced and complex modern technologies that can be used to mimic natural human behaviour.

And you probably see where I’m heading with this; Yes! This is ideal for applying to a chatbot, where a user will then have a very similar experience when talking to a chatbot as when to a human. These technologies are incredibly handy for chatbots as they give it so much more
possibilities. For example: Natural Language Processing allows computer programs to understand spoken and written text the same way as humans do. This would mean that instead of making use of very short keywords, a user can now type “normal” sentences containing much more information, and then the program can use all this information to respond to the needs of the user.

This might sound rather confusing as we, humans, think that our speech and texts are “simple” and “normal”, so it should not be too hard to process the information. But on the contrary, human speech is rather difficult to dissect and process in the different bits of information and background information.

AI and Machine Learning allow the chatbot to learn to give the right answer to the right question on it’s own, with a little help from us humans of course since chatbots won’t become intelligent overnight, but rather need quite substantial training before being able to stretch their wings and server user needs on the world wide web. The more input the chatbot receives, the better it gets at giving the right answer.

Where are we heading

Since no one can predict the future, I certainly won’t try that here either. But we can observe the trends in chatbots and technology in general to form an idea where the future will take us.

Some very recent trends and applications are these so called “fluid assistants”. They make use of cutting edge technologies and are trying to solve some very specific problems. Companies that are specializing themselves in these fluid assistants are for example Ada, a company that focuses on improving health by developing a health-care chatbot service. Ada helps people understand, manage, and get care for their symptoms with trusted medical expertise in minutes. It essentially checks (and interprets) your symptoms without a doctor’s appointment. Ada is 24/7 available, and can be the health profile Ada makes can be of use for your personal doctor.

We can also see a lot of development in making chatbots a customer engagement and marketing tool, as chatbots not only cut down on company costs, but also generate a lot of new revenue!

To close with a quote by Alan Turing, the British computer scientist who not only cracked the Enigma code that shortened the Second World War by many years, but also was the pioneer on computer science who started toiling on the thought whether or not computers would be able to think:

“It is possible to invent a single machine which can be used to compute any computable sequence.”