Bots are very trendy in world of technology right now. Under the right circumstances, they can be very useful and practical especially with the great shift on how we consume information in the modern world. Nowadays messaging apps are very common for personal use and work and companies are working hard to utilize them to better serve their customers. Building bots has become very easy with services like Microsoft Bot Framework which offers a great deal of features and capabilities. This is all very nice until you need to communicate with the bot via natural language. Unless your team has an expert on the field and a ton of data to train a custom build language understanding system you are out of luck.
Here is where LUIS.AI comes to help. It stands for Language Understanding Intelligent Service and it's build by Microsoft and it is part of Microsoft Cognitive Services and integrates seamlessly with multiple services.
Image 1. LUIS integrates very well with different services (source: luis.ai)
First thing to know when using conversational agents (LUIS.AI) is that they need to understand what actions the users want to perform (the intent, ex. Book a hotel room) and collect the data needed to perform that action (the entities, ex. room size, price etc.) in order to perform an action. At every interaction, based on the history (the state) and the current user input, it should either request information, ask for confirmation or perform some kind of action.
LUIS.AI needs to be trained by the developer before being able to determine intent and entities. The training is done by providing LUIS with examples of what user would write as input and mark the tag words which will serve as entities. Once it is trained with enough examples, the algorithms should be able to provide significant results on inputs that are similar to the examples.
The best thing about LUIS is that it integrates flawlessly with the Bot Framework. With Bot Framework business logic and conversation flow is written inside callbacks that listen to events; when LUIS recognizes intents and entities, it emits the related events and callbacks get executed. There is quite some work to be done by the developer, but it’s the best compromise between flexibility and not having to build a framework from scratch.
10000 free transactions per month
$0.75 per 1000 transactions
Up to 50 transactions per second