Chatbots are programmed to simulate human conversation and exhibit intelligent behavior that is equivalent to that of a human.
Mary Beth Moore | AI and Language Analytics Strategist |SAS
Introduction: Chatbots/talkbots/virtual assistants are a form of conversational AI, that is designed to simplify human interaction with a machine (computers, phones, tablets, etc.) Through the use of chatbots, a machine can recognize, understand and respond to human inputs through spoken or written language. The very first chatbot was developed at MIT (Massachusetts Institute of Technology) in 1966. The chatbot was called ELIZA. One of a kind and often called the mother of all chatbots, ELIZA answered elementary decision tree questions.
Through two key developments namely, emergence of messenger applications and extensive development in Artificial Intelligence and Machine Learning humans have recognized and developed extensive use-cases of chatbots.
Ranging in complexity from information gathering, retrieval and learning capabilities that provide rigorous analysis and tailored suggestions the use-cases of chatbots are indefinite. Some of which are discussed below:
Customer service chatbots: That answer frequently asked questions or resolve customer queries. Repeatedly used for resolving customer service issues and when a customer, for instance, wants to check the delivery time of their package.
E-commerce chatbots: Retail companies and telecommunication providers that use e-commerce platforms often use chatbots as an additional interaction channel with their customers. E-commerce bots have experienced exponential growth especially this year (2020) due to COVID-19 which placed physical restrictions on customers. These bots support customers with their queries and escort them through a process flow to complete requests or transactions.
Virtual assistant chatbots: Siri — the first voice assistant was soon followed by other virtual assistants like Cortana and Alexa who have risen immensely in popularity as their benefits are “one-click” away and easily embedded into our routine life. These are frequently used for quick information retrieval, scheduling, and interacting with smart home features.
Other use-cases: Finance, IT help-desk assistants, travel optimization bots, Inventory Management, and healthcare bots.
In current times, not merely using but also developing chatbots have become seamlessly easy. Chatbots can be developed for small businesses, stores, and even individuals. Two ways of having your “own personal-virtual assistant” are either writing numerous lines of code or through a few clicks on tools like Watson Assistance or Azure Bot Service to name a few.
Underneath is a snippet of the code that can be written to develop your own chatbot for a flower shop. I named it DAISY (Well, since you can build your own personal assistant, you can name it as well.)!!
Chatbots dialog skills are comprised of three components:
Intent: capturing the user goal | Entities: Allow users to capture specific information in user input | Dialog: The tree of nodes.
On contrary to compiling the code, chatbots can be developed through numerous platforms such as Watson Assistant (formerly Watson Conversation) and can be leveraged for its AI capabilities. “Watson Assistant remains a remarkable piece of technology that will allow anyone, not just programmers, to build smart chatbots and conversational interfaces” — IBM.
Watson Assistant is a converstaional AI platform that enables organizations and businesses to provide their customers with fast, accurate, and straightforward solutions/answers to their questions. Watson Assistant, along with providing the most reliable service and solutions, also, reduces the cost of customer interactions, bringing the organization’s agents in the picture for helping with complex use-cases — not repetitive customer queries.
Below is a brief overview of how IBM Watson Assistant can be leveraged for developing your own chatbot:
2. Adding skills: This means preparing the machine to respond to the customer (or the bot user) queries.
– Adding skills to your assistant and deploying to your channels.
3. Creating Intents: An intent comprises a collection of user statements that have the same meaning. By creating an intent, you can train your virtual assistant to comprehend various ways in which a user expresses their goal.
Example: Defining a “greeting intent” by entering numerous ways of greeting such as, hello, hi, good morning, and so on.
4. Create, modify, and delete entities: Entities in bots help recognize and capture specific pieces of information in the user input.
Example: In our flower shop, there are significant chances that the customers would like to know about store operating hours, locations, and so on. Thus, in our bot, we will have entities like @location, @operating_hours, etc.
5. Dialog: In the process of chatbot development, the dialog acts as a bridge as it enables a human user to interact with the computer system. Here, the chatbot builder inserts different dialog statements (responses) which will pops-up when someone is interacting with the bot.
For example: When a user first opens up the chatbot a welcome message can be inserted for the user. Like the welcome message, I inserted for my chatbot.
6. Customizing the bot: The purpose of chatbots is to support customers or users and, to achieve that efficiently it is essential for the chatbot to have the ability to resolve domain-specific questions. Like, giving information that is related to a specific location or occasion (for the flower-shop chatbot). This can be achieved by creating parent-child nodes.
7. Deploying: The personal chatbot can be deployed for others to see and interact with through IBM Watson Assistant which comes with a free allowance of 10,000 API calls per month.
8. Deploying the chatbot to your own website: Not just for testing but once the IBM Watson Assistant has been developed and customized, you can also deploy and activate the Watson Assistant plugin to your own website.
Developing chatbots and customizing them as per the requirements is easy and an efficient way of operating businesses. Chatbots are among the most visible applications of Artificial Intelligence in the current time. An increasing amount of use-cases of chatbots can be observed all over the spectrum as companies push more towards AI. Developing and optimizing chatbots is seen as a part of a broader push for innovation in artificial intelligence.
1. Building your own point-and-click chatbot:
IBM Watson Assistant: https://www.ibm.com/cloud/watson-assistant/
Azure Bot Service: https://azure.microsoft.com/en-us/services/bot-service/
2. My chatbot: https://webchat.global.assistant.watson.cloud.ibm.com/preview.html?
3. Learning Resources: