Rasa 3.x Framework Guru-AI assistant creation platform

Example

Rasa's NLU pipeline processes user input to extract intents and entities. For instance, in a food delivery bot, when a user says, 'I want to order a pizza,' the NLU model detects the 'order_food' intent and recognizes 'pizza' as an entity.

Scenario

In a customer service scenario, Rasa’s NLU can handle multiple intents, such as 'check_balance' or 'transfer_money', based on user utterances. The extracted intents and entities help trigger the appropriate actions, making the interaction smooth and relevant to the user's needs.

Example

Rasa’s dialogue management relies on policies like TEDPolicy and RulePolicy to control how conversations flow. For example, in a healthcare assistant, the bot may ask follow-up questions based on previous user responses to gather more information, such as symptoms or health conditions.

Scenario

In a technical support chatbot, Rasa’s dialogue manager can guide users through troubleshooting steps, dynamically adapting based on the user's responses, and escalating the conversation to a live agent if needed.

Example

Rasa allows developers to write custom actions that connect the chatbot with external APIs or databases. For instance, a travel bot might use a custom action to retrieve flight availability from an airline's API when a user asks about flight options.

Scenario

In a financial services chatbot, a custom action could be used to check a user's account balance by querying a backend system, then returning the data in a user-friendly format. This extends the assistant’s capabilities beyond basic question-answering to more dynamic interactions.