What is a Chatbot?
Remember the days when you used to get your answers to daily life problems through simple conversations – Chatbots are a throwback to those simpler times!
A Chatbot is a service, leveraging artificial intelligence technologies, that you interact with via a chat interface. While a Chatbot can be functional and solve business problems without the need for human intervention, it can also engage and retain site visitors. An engaged site visitor is more likely to convert into any purchasing activity.
Chatbots can be programmed to carry out automated actions. Chatbots can both initiate actions as well as respond to requests from other users.
How do Chatbots Work?
Chatbots understand your messages through a Natural Language Processing (NLP) engine, such as Microsoft Cognitive Services or Amazon Lex or IBM Watson. The point of NLP is not to interpret sentences word-for-word but to extract the ‘intent’ behind the messages. For example, the intent of question “how will be weather tomorrow?” could be “weather forecast.” Keeping context is the Chatbot’s ability to remember and combine intents across several sentences in a conversation. Let’s say the Chatbot in the above example responded with “It’s sunny with a couple of light showers during the day.” If then you ask, “How many degrees?” the Chatbot would assume the context of weather and output today’s temperature.
Where are Chatbots Used?
- E-commerce Chatbots enable buying of goods and services.
- Content Chatbots share relevant content with you (e.g., news, weather).
- Watcher Chatbots alert you to specific events (e.g., your flight is delayed, or your car is due for service).
- Banking and trading Chatbots provide financial services.
- Workflow Chatbots automate business workflows in sales, HR, operations, admin.
- IoT Chatbots connect us to our smart homes, cars, and devices.
- Personal Assistant Chatbots manages communications, scheduling, and other personal tasks.
Can Chatbots Integrate with Backend Systems?
Chatbots are all about answering questions. These questions can be asked in a simple natural language manner and do not require any scripting or programming. Chatbots make the whole experiences of getting answers by asking a series of questions fun, interactive and time-saving. But what goes on behind the scenes? What multitude of technologies, the outcome of decades of research & development efforts, come together seamlessly to work behind the scenes to get your answers?
Assume that a Support Rep wants to look up a customer’s order status. For this, the Rep would have to ask some simple conversational commands. The Rep would be presented with a set of valid search options like, lookup a customer, review a customer’s most recent order, etc. After a series of questions and answers, the Rep would be presented with the customer’s order status and any other information that he/she may be interested in.
Here is the logic flow of such an Information Bot:
- The employee starts the Information Bot.
- Azure Active Directory validates the Rep’s identity.
- The Rep can ask the bot what type of queries are supported.
- Cognitive Services returns a series of answers built with the QnA Maker.
- The Rep defines a valid query.
- The Bot submits the query to Azure Search which returns information about the application data.
- Application insights gather runtime telemetry to help developers with Bot performance and usage.