Salesforce Einstein Bot Looks Like Your Training Job Didnt Complete Try Again
Have you lot ever been on the phone with a service provider and had to go through 800 menu options and routing configurations to go to a person? Or been chatting with a chat bot who clearly just doesn't understand what you demand to exercise and you only really want to talk to a person? Aye, me too. I discover myself pounding on the 0 or typing "talk to an agent", "talk to a person", "human"… trying not to be awful just merely wanting to get somewhere. You may have guessed already, simply I'm writing about Salesforce Chat again… and more importantly, Einstein Bots. Out of the box, your bot isn't that smart. Its a set of pre-programmed dialogues that do specific deportment similar send messages or run rules and deportment in the groundwork. If I but outset typing random things like "send me to your leader" the bot is not going to understand what I'm saying. Luckily, Einstein Bots handle defoliation by routing to an agent, but we tin handle it a whole lot better by using Intents.
Kickoff simple and iterate
Similar I said earlier, out of the box, your bot is a series of menus that end users tin can select from. This is chosen a menu-based bot and should exist the starting point for every bot you ever make probably. In one case yous've mastered that, you tin can start adding some intelligence using Intents. Your bot uses tongue processing to attempt and effigy out what your customer is saying, and that is washed via an Intent model. I highly recommend checking out this documentation that goes through the stages you should accept with your bot. It recommends starting with a carte du jour based bot, and so graduating to a hybrid bot that uses both menus and Intents using what is called an Utterance, which is just the input from your end user. Since we just launched our bot, we don't have a ton of utterances yet. The beginning Intent I want to start with, is the "Transfer to an Agent" Intent. I'grand starting with this just considering of my past experiences with bots.
In order to use Einstein to train a existent model, you need to accept at to the lowest degree xx utterances that yous've categorized. This means a lot of watching your bot, a lot of categorizing utterances, and a lot of iteration. Until yous have enough to turn on Einstein, your bot volition match utterances exactly upward to 20 utterances. After the 20, you lot can turn on Einstein to decide based on like words. If a customer inputs text that matches ane of the utterances I accept categorized for this dialogue, and then that dialogue volition be started.
Hither are the utterances I've set so far based on just what I think people would type into the chat window to go to a person.
So you'll encounter hither on row 3 I have the word "transfer". This means that if I type "Transfer" into the chat window, my bot will recognize that and begin the "Transfer to Agent" dialogue.
In your Model Manager, under Bot Grooming, y'all can run into all of the utterances that are existence nerveless and offset categorizing them.
To showtime classifying your utterances, simply click on the little carte to the correct of the row you desire to allocate and click "Reclassify". You'll likewise find an selection to "Ignore". In the instance below, I'thou going to classify the row that says "main bill of fare" as my Primary Carte du jour intent. Y'all'll see that it'due south super easy!
I'm super excited to keep an heart on this bot and get-go analyzing the utterances we're collecting. By looking at these utterances, you can besides get a better idea for what your customers might be expecting your bot to practice. For example, my testers were speaking in dissimilar languages – maybe I need to accept bots that speak different languages. They were as well trying to inquire very specific questions that would be like shooting fish in a barrel to create a dialogue to respond. I think that making fourth dimension to review utterances at least once a week will be an important part of bot maintenance. I'm but guessing on once a week… I don't expect conversation to become a total time job, but it volition require ongoing grooming and analysis for sure. I'll keep you posted on how much time and effort goes into the bot, and will write some other post once I'm able to enable Einstein to help understand our customers!
Source: https://forceforfun.com/2021/08/03/before-you-put-the-einstein-in-your-einstein-bot/
0 Response to "Salesforce Einstein Bot Looks Like Your Training Job Didnt Complete Try Again"
Post a Comment