Watson conversation better than Watson Natural language classifier? -


i created classifier in nlc after creating number of classes(intents) , few examples each class. classifier accuracy poor.

when used same training data in watson conversation service , tested it, intent identification accuracy good.

as understand conversation service uses nlc, why there such big difference in performance?

watson conversation not use nlc. have 2 different learning models. conversation works better natural language classification, while nlc text in general.

nlc takes longer build it's model vs conversation. take conversation few minutes, can take 30 minutes or more nlc.

lastly nlc uses called relative accuracy, while conversation uses absolute confidence.

to explain, imagine have 2 intents "cars" + "trains".

if ask nlc question, tell car or train, if question "what elephant?". confidences added add 100%.

conversation on other hand can understand question might not related @ trained on. can tell elephant not car, nor train.


Comments

Popular posts from this blog

commonjs - How to write a typescript definition file for a node module that exports a function? -

openid - Okta: Failed to get authorization code through API call -

ios - Change Storyboard View using Seague -