Chatbots are all the rage. They exist for social media, websites, stores, and even business to business conversations. Chances are you’ve spoken to one, but may not have even realized it. Chatbots are taking over the Internet, saving companies hundreds of hours and manpower. Explore the world of chatbots, how they came to be, and what they are capable of.
Want more information?
To put it simply, chatbots are computer applications powered by artificial intelligence to communicate with you. They are designed to mimic natural human conversation via an online chat interface, SMS, and sometimes, even voice chat. Despite being designed for human consumption, chatbots are also capable of communicating and gathering information from other chatbots.
Much like how it sounds, artificial intelligence is when a computer simulates humanlike intelligence. For a computer to be considered intelligent, it must be capable of learning, reasoning, recognition, and correcting errors. It is not, however, the Hollywood interpretation of robots that are virtually human (Such as Marvin the Paranoid Android from Hitchhiker’s Guide to the Galaxy or C-3PO from Star Wars). That is still purely science fiction and we are quite a long way off; it’s very likely to not happen within our lifetime. In order for a machine to act or behave like a human, they require a great deal of information about many different situations, items, categories, and relations for them to connect each occurrence to one another in a method that is called knowledge engineering. It includes groups of complex information compiled into a knowledge base and linked through building, maintaining, and using the information.
Strong AI: Strong AI is when the machine involved is developed to simulate human behavior and cognitive abilities, as well as the ability to reason and learn. This is what people typically think of when they think of AI: robots that can respond to questions, solve problems, or perform complex tasks. This type of AI is adaptive and has opportunities to expand.
Weak AI: Weak AI is when a machine is designed to carry out a specific task – such as working an assembly line, or playing chess against a master chess player. It is incapable of self-improvement, so it will only be good at what it’s programmed to do, nothing more.
Machine learning is a discipline of computer science that explores the construction and study of algorithms that can learn from data. Such algorithms operate by building a model based on inputs and using that to make predictions or decisions, rather than following only explicitly programmed instructions. There are several different processes that machines are capable of understanding. Some of the popular machine learning methods include supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning.
- Supervised Learning: Where data is labeled and described to the machine in order for it to detect and sort the new information.
- Semi-supervised Learning: There is a small amount of labeled data coupled with a large amount of unlabeled data to train the machine.
- Unsupervised Learning: Information is not specified or labeled in any way, but may be sorted in reference to their similarities or differences.
- Reinforcement Learning: The machine is given a specific goal to follow through with, followed by a review and providing feedback, whether positive or negative.
Have you ever put any thought into how simple it is for humans to understand language? Just think: in everyday conversation, we convey our thoughts through a series of sounds that make up a language. How our brains are capable of translating so much unstructured data into useful information can be unfathomable. Language does not come as naturally for machines; it takes a considerable amount of programming for them to understand even basic statements. So how do they do it? With NLP. Natural Language Processing (NLP) is a concept that is still very abstract to most–even those that are in the IT industry can struggle with the idea. NLP is a type of “program” designed for computers to read, analyze, understand, and derive meaning from natural human languages in a way that is useful. It is used to analyze strings of text to decipher its meaning and intent. Essentially, NLP is a way to help machines understand human language. It is far more complex than that, however; think about how many ways a single sentence can be spoken and understood by the recipient. Here is a brief example of things that all are asking the same thing, but can be spoken completely different:
- What’s the weather report?
- Do I need an umbrella?
- Is it raining?
Natural Language Processing will take the intention of the request and find the most appropriate answer to your question in the same way another human would.
Similar to how a child learns how to communicate, deep learning is when a bot is given an artificial neural network with a loose set of rules for it to abide by. While not perfect, it is a way to experiment with different algorithms to create a bot that suits your needs. Another benefit of deep learning is that once completed, there is no need to reprogram as the machine learns on its own. If it makes a mistake, you provide corrective action for it to learn. Not very many bots are currently utilizing deep learning due to the difficulty of creating an artificial neural network, but many companies that are actively researching it.
An English computer scientist by the name of Alan Turing developed the Turing Test in 1950. It is a controversial test designed to test the ability of a machine to exhibit intelligent, humanlike behavior. To pass the test, the machine’s replies would have to be indistinguishable from a human in a five-minute test. Turing predicted that by the year 2000, machines would be able to fool 30% of human judges, but this feat took much longer than anticipated. As of 2017 no machines have truly passed the Turing Test, but there have been claims that a chatterbot by the name of Eugene Goostman passed the test in 2014 by fooling 33% of the judges involved into believing he was a human being. The use of the Turing Test has been very controversial. The test was not intended to test a machine’s intelligence, only that a machine behaves and acts like a human being would when faced with the same situations. As not all human behavior is intelligent, the Turing Test would also test for behaviors like lying, making typing mistakes, or responding to insults. Another reason that intelligence is omitted from the test is that machines are far more likely to solve complex mathematical problems that would be impossible (or nearly impossible) for a human to solve.
The idea of creating a machine that had humanlike thought processes has been around for centuries. Scientists, philosophers, and even sculptors were fascinated by the idea of a humanistic automaton. Author Samuel Butler first wrote the idea of a mechanical consciousness in his 1872 science fiction novel, Erewhon. Despite the interest and fascination on the subject, it was not until 1966 when any form of artificial intelligence really took form: ELIZA.
Generally recognized as the first actual chatbot, ELIZA was developed by Joseph Weizenbaum. Named after Eliza Doolittle, a workingclass character in the play Pygmalion, ELIZA was meant to emulate a Rogerian psychotherapist. It was capable of answering basic questions and asking for users to elaborate on their discussions. ELIZA was a groundbreaking invention despite its lack of “real” intelligence. Despite only being capable of using pattern matching and substitution methodology to form new sentences, ELIZA would receive an overwhelming response of positivity for its “human-like” conversations; even Weizenbaum’s secretary was taken by the program. It was the inspiration for dozens of chatbots.
One of the chatbots inspired by ELIZA was A.L.I.C.E (Artificial Linguistic Internet Computer Entity), also known as Alicebot. Alicebot was the brainchild of robotics professor Richard Wallace, officially going live in 1995. Alicebot uses AIML (Artificial Intelligence Markup Language), a programming language that describes a class of data objects (called AIML objects) and partially describes the behavior of computer programs that process them. Essentially, it will have a list of categories to discuss with the user – most bots have around 45,000 different topics to choose from – and choose the most “accurate” response for the discussion.
Many older millennials will remember SmarterChild, a chatbot designed for use with various SMS platforms in the early 00s. SmarterChild was revolutionary in the fact that it was capable of providing the users chatting with news reports, weather updates, movie showings, and more. In the time that it was active, it had chatted with over 30 million users on AOL Instant Messenger and MSN Messenger alone. SmarterChild was an enormous success, spawning multiple branded spinoffs and taking the internet by storm. It was so successful that the company that developed SmarterChild, ActiveBuddy, filed a patent for it in 2002. Unfortunately, ActiveBuddy (and at that point, Colloquis) only lasted a few more years after, Microsoft acquiring it in 2007 and subsequently decommissioning the business.
In 2011, history was made when a machine contestant was on a game show. Squaring off against two former champions – one of which that had a 74 game winning streak – IBM’s super computer, Watson, was capable of understanding idioms, riddles, and nuances well enough to score over $50,000 higher than either of the contestants, resulting in the win of a $1 million prize. Of course, Watson wasn’t always correct, sometimes giving absolutely bizarre responses to questions asked. An example being, “Its largest airport is named for a World War II hero; its second largest, for a World War II battle.” The response? “What is Toronto?????” – The correct answer was Chicago. Since appearing on Jeopardy, Watson has expanded its capabilities, its processing capabilities, and decreased its size substantially (going from the size of a large bedroom to that of a mere three pizza boxes). Now considered a platform rather than a Jeopardy! contestant, Watson is paving the way for enterprises to create more believable virtual assistants, create more accurate reports, and perform intense research.
You can consider a chatbot as a type of app without the familiar user interface. There are inputs and outputs, a database, APIs, and a great deal of code involved. The biggest differentiator between an app and a chatbot, however, is the addition of a Natural Language Processing (NLP) engine. As human language is often imprecise, and, depending on the subject and language in question, one word can mean 500 different things (the English word “run” currently holds the record at 645 different definitions). That is exactly why the NLP engine is the key differentiator. Made up of thousands of different libraries of definitions, meanings, and classifications, the engine will identify relevant pieces of information provided by the user and determine the correct output in response. The engine uses tasks such as named entity recognition and tokenization to discern exactly what it is the user is wanting.
Chatbots are most well known for being customer service agents, able to handle a majority of a customer’s questions without the need for human intervention. Did you know they are much more capable of just being support? Chatbots can help with countless things in today’s modern age. We have bots that can reserve hotel rooms, provide companionship, set up meetings between a group of people, send money between users, and more. One of the most recent chatbots to take the world by storm is a free “lawyer bot”. While it won’t be replacing a human lawyer any time soon, it is capable of helping you create a case to appeal parking tickets, get reimbursed for travel complications, or provide answers to tricky legal questions. Essentially, chatbots can do just about anything you want them to do. It all just depends on their programming. Some are even capable of reading emojis and pictures, responding to them as a normal human would.
Examples of Different Bots
- Reddit’s AutoTLDR Bot: AutoTLDR is for those users who like to keep up with the news but don’t have the time to read an entire article. It takes brief snippits of the article and creates a 450-700 character summary, posting it on the forums along with a link to the full-length article.
- Mezi, Your Personal Travel Bot: Tell the bot when and where you want to go and where you’re flying out of and it’ll search and book a flight and hotel for you.
- Food Delivery Bots
- A.I. ♥: A chatbot “doctor” that is capable of assisting with the basic diagnosis of common ailments, such as bronchitis, a cold, a sprained ankle, or fever, by asking targeted questions.
- Poncho: Poncho is a “weather cat” chatbot that originally started out simply doing morning weather updates. It has since evolved into what the creators call “a bot you can be friends with,” now able to provide dating advice, make jokes, and more.
Whether your business is a business-to-business (B2B) or business-to-customer (B2C), chatbots are capable of improving customer service, automating repetitive tasks, and freeing up significant amounts of time. Here are only a few of the benefits your business can see with a chatbot:
- Internal Communications: Improve communication between users, departments, and locations by an automated bot being able to announce news, changes, or updates.
- Assistants: Focused on the more administrative tasks, Chatbot Assistants are helping you document your timesheets, keeping you on track with all of your assigned tasks, or even reminding you of meetings or events coming up.
- Knowledge Base: Get fast, easy access to documents or information without having to search for it in an expansive library.
- Providing Insight: Things like stock prices, your website’s page views, or amount of contacts created the previous day are only a few examples of an InsightBot.
Of course, not all the benefits are strictly internal. There are many reasons that a B2B should consider an externally facing chatbot alongside an internal one.
- Answer Customer Questions: A potential or existing customer may not be ready (or want) to speak to an actual salesperson, so a chatbot could be a high-level assistant, able to answer questions about products or offer advice.
- Up-to-Date Tracking: Keep a log of communications that are happening between customers and the bot. It gives you direct insight into what customers are looking for, what they’re asking, and if there’s anything you need to improve.
How Do Chatbots Help B2Cs?
While still capable of receiving the same benefits of B2Bs (and vice versa), B2Cs have some especially good advantages of utilizing a chatbot for their business. Some examples are as follows:
- Instant Availability: The fact that chatbots are online 24/7 is a significant advantage when you have a product to sell.
- Ecommerce: Promote sales, make suggestions based on previous purchases, manage returns, and even take payment with an ecommerce bot.
- Customer Service: Instead of having an entire call center swamped with calls about everything under the sun, there can be a chatbot to front the customers and answer basic questions, or escalate to a live agent as necessary.
- Transactional: Some banks are even using chatbots to help you do your online banking without even having to access the bank’s website. Instead, you just chat with a bot to get whatever information you need out of it.
Want more information about chatbots? Fill out the form and one of our experts will reach out to you shortly. Toll-free 1 (844) 425-8425
By submitting this form, you authorize V-Soft Consulting to contact you periodically. We will never give your contact information to a third party.