– Classic workflow –
Usually, Chatbots are build from intents and actions.
An agent is trained to match user inputs to a predefined intent. Then the intent triggers an action, for instance by calling an external Webhook.
– An example –
1) User input: “Hi, I’m xyz@abcd.com, can you tell me the status of my order XYZ? It looks like stuck.”
2) Bot: “Of course. Your order XYZ is waiting for the bank’s approval. I’ve asked a sales rep to call you back.”
– Explanation –
In this example, the input is matched to the intent “order status”. The parameters {client_email} and {order_number} are extracted and used as input to a call to WooCommerce orders API to get the status and staff comments.
– New kid on the block –
OpenAI function call is doing exactly that: matching a prompt to a list of functions with parameters.
The major difference: the call to the function is not triggered by the agent. Instead, OpenAI returns the function and parameters to call, and the API caller calls it. Then the caller appends the result to the initial prompt, and OpenAI will return another function to call (this can be done several times), or a summary of the result.
– WooCommerce –
Of course, the API caller will be WPSOLR, and the functions called will depend on the needs: order status, recipes generated by OpenAI from a list of products retrieved by Weaviate, and so on.
This is a new concept, where a bot will mix questions, search, and actions to ease the client intents.
Search until now was rather passive. By mixing it with conversational and generative AI, search will become active and produce things that are limitless.
OpenAI function calling
WPSOLR + WooCommerce + ChatBot(?): https://wpsolr.com
#wpsolr #weaviate #chatbot #woocommerce
