Intents in CXone 机器人生成器
Intents are the core meaning behind the contact's words; the actual result or goal the contact wants to communicate or achieve. Follow these best practices when creating intents:
- Use real data. Don't use tools that auto-generate data and claim to train your bot
代替现场人工坐席处理客户交互的软件应用程序。 faster. They often produce examples that don't reflect what contacts really say. They also can result in a bot that loses its ability to generalize. Over time, the bot reaches a point where it only recognizes phrases it's seen before.
- Don't use the same training examples for more than one intent. Intents aren’t always clear-cut. Two user goals might seem different but start to gather similar examples over time. When this happens, see if you can merge similar intents into a more general category.
- Always include an out-of-scope intent. Out-of-scope intents cover anything the user might say that’s outside of the bot’s domain. They allow you to recover the conversation and often result in improved performance.
Multi-Intents
Bots 代替现场人工坐席处理客户交互的软件应用程序。 built with CXone 机器人生成器 can identify both single and multiple intents
联系人所说/输入内容背后的含义或目的;联系人想要沟通或完成的内容 in a contact's message.
Multi-intents can be simple. They can also be more complicated and specific to your business. For example, a bank might need to create a multi-intent to handle a message like, "Please show me my balance and transfer $200 to my savings account."
If you create a multi-intent from two existing single intents, you don't have to provide many examples. Your bot can recognize the single intents in the message and handle them as a multi-intent.
Only use multi-intents when they are really necessary to the natural flow of conversation. Too many multi-intents can make your bot too complicated to manage easily.
Intents Tab
The Intents tab inside NLU displays the Inbox and a list of all your intents. From this tab, you can create intents and add examples to existing intents. You can also visualize how much each intent is used in your training data and dialogues (stories and rules).
- The number on each intent represents how many training examples there are for that intent. These numbers are color-coded to indicate if the training for the intent is sufficient.
- Red: Intent is used up to seven times.
- Orange: Intent is used between eight and fourteen times.
- Green: Intent is used more than fifteen times.
- The yellow-orange triangle
indicates there are no dialogues (stories and rules) that include this intent.
- Each intent has a friendly name. This is what the bot will use if it needs to directly refer to the intent to the customer, such as in advanced NLU fallback.
-
You can hide individual intents. This excludes them from training until they're ready. Hover over an intent, click Options
, then click Hide. When hidden, the intent is visible to you for editing, but your bot does not learn from it. It is also displayed with the Hidden
icon. When the intent is ready for training, click Options
, then click Show, then click Train and Stage.