Entity

Both Dialogflow ES and Dialogflow CX support entities and they also behave similarly. An entity represents a predefined categoy of objects and the basic idea is borrowed from the field of Natural Language Processing. Named entity recognition (NER)—also called entity chunking or entity extraction—is a component of natural language processing (NLP) that identifies predefined categories…

Event

Events are used in both Dialogflow ES and Dialogflow CX as triggers – they trigger something like an intent or an intent transition route – which fire without any user utterance. The best example is the WELCOME event in ES which has a very specific role – it will automatically fire the intent which contains…

Intent

Both Dialogflow ES and Dialogflow CX have the concept of intents, but they don’t have the same behavior. An intent represents what the user wants. “Switch on the lights” “Order a pizza” etc. In the case of Dialogflow ES, the bot’s response is added into the intent itself. But in Dialogflow CX, an intent only…

Parameter

Both ES and CX support the concept of parameters. If entities are the categories or types of proper nouns, parameters are the actual values. For example, “city” would be an entity type, and Dallas would be the parameter (entity value). Dialogflow ES and CX allow you to extract parameter values from a user utterance. For…

Scope

Dialogflow CX scope helps you decide what set of actions CX can do based on the current state of the conversation. For example, only intent transition routes in the current page can fire (be invoked), and that’s because intent transition routes in the other pages are not in scope. For a more detailed discussion, check…

Simulator

You can use the Simulator in both Dialogflow ES and Dialogflow CX to test your conversation without incurring any charges. Here is what the simulator looks like in Dialogflow ES: And this is what the Simulator looks like in Dialogflow CX: Given that Dialogflow CX is based on state machines, it provides a lot more…

Utterance

An utterance is simply what the user says to the bot. Both ES and CX uses the phrase “training phrases” to specify user utterances. Here is what it looks like in ES: