Welcome to the Rocketbots Academy. Today, we’ll show you how to build your own basic chatbot. But first what is a chatbot? A chatbot is an artificial intelligence powered program that can answer questions automatically. You’ll tend to find chatbots on some of the most popular messaging apps like Facebook Messenger or WeChat Chat, bots are a great way to engage customers and collect information that’s valuable to you Let’s take a look at a great use case for a chatbot Today, we’ll be building a chat bot for Sally who owns and manages a small fruit shop Lately, she’s been getting a lot of enquiries over Facebook, but she’s busy at the shop and can’t spend much time responding to these messages However, she does notice that the enquiries are mostly similar. There are also easy questions to answer Sally then decides that she should use a chatbot to tackle.
These frequently asked questions She then begins to collect the most common questions. These questions are called expressions in chat bot development expressions represent a user’s desire or aim She notices that more than a few of these expressions are essentially asking the same thing With questions, like are you open on Sundays? Are you open on the weekend? And so on we can see that the user wants to get an idea of what our store hours are like We can continue with this on other types of expressions Let’s group these expressions by the aim the user has We call this grouping an intention. Sally then continues to group the other questions into their corresponding intentions We now see that our three most common intentions are for opening hours products and location These are the intents our chatbot should tackle for Sally Now all we need to do is provide a response that satisfies each intention Once this is done sally has now planned the purpose of her chat bot and is ready to build it and deploy on speak messenger Now that she knows the chatbot’s job Sally’s decided to use two free tools Dialogflow to build the chat bot.
It also acts as the NLP or natural language processor for the chat bot. It’s essentially the brain She also wants to use the Rocketbots platform to deploy and manage it First she goes the Dialogflow and makes an account once at the Dialogflow dashboard click add agent to the top left Give the agent a name and make sure the agent is in the right language and time zone click create Once the agent is created it’s time to create our first intent Click create intent at the top right and give the name opening hours Click Add training phrases here. You’ll be able to input our expressions Once finished we can scroll down and click Add response. Let’s add the store hours here Remembering to click Save for the intent at the top right repeat the process for each of the other two intents Once finished the chat bot is at its most basic level and ready for deployment on Facebook At the Rocketbots platform dashboard clicking the plus sign next to the bot list You will be brought to a window that lets us connect our agent from Dialogflow Give the bottom name once again now, we’ll select Facebook messenger as our channel click continue again And you’ll be prompted to sign in to your Facebook account.
You may have to give Facebook permission to connect Now select your Facebook page and click continue We need to provide her agent tokens from Dialgoflow to continue Go back to dialog flow and click on the COG at the top left beside your agent name Scroll down until you find your tokens choose version 1 API and copy/paste the tokens below into their respective fields on the Rocketbots platform Click continue finally the chatbot is connected click finish the deploy your chatbot.
Read More: ChatBot AI