intents Conditions
# 🤝help
w
I have my intent file (intents_policy) and I have the node that uses the intent and it works great if I use an word/sentence in the file. My question is there is a way to change the expression from : *"intent is : intents_policy*" to something like intent contains : intents_policy ? Using a contains or a like here would be more useful in my scenario where if I had to add in every sentence that a user may ask would be a lot more complex (lots of abbreviations possible). If I have to do this via javascript or typescript I am good to do so, but where would I find this documentation? Thanks! Tony
b
hm, can you give me some more details about what you're trying to make easier by doing this? or what you'd like to avoid happening?
w
Robert- Thanks for getting back to me so quickly. One path of the chatbot will take is to allow the user to ask questions regarding their "policy". The policy contains topics in regards to items such as Home Finding, Temporary Living, etc. I want to be able to send this chat bot to search the policy if someone uses specific words or phrase: Home Finding, HF, Temporary Living, Temp Living, TL, etc. There are 10 different categories to the policy all which can be abbreviated multiple ways. And to put in full sentences the user may ask would become very cumbersome. My idea is to give the user a single choice option to start, but if they just type something I would like to handle it rather than just loop and force the user to select an option (could do it this way if needed). So if I could use a "contains" based on the intent_document it would be quick and easy to send them to the policy tree. Does that make more sense? Tony
b
intents work with NLU, so you don't need to include all of the exact sentences your user might say
if you include, say, 10-20 possible examples of what a user might ask for a certain "path", then your bot will be able to interpret variations of those sentences, even if they're not included verbatim in your examples (these are called utterances)
there's also a simpler feature - AI Transition - which "listens" to what a user has said, and adjusts the path accordingly, based on a given "category" that you define
w
Great! Thanks for the information. I will definitely check out the AI Transitions and play around some more with the intents and the bot interpretations. I appreciate the quick response and the information. Keep up the good work, your video's have been a huge help!! Tony