1. Home
  2. Kore.ai Conversational Platform
  3. Bot Builder Tool
  4. Natural Language Processing (NLP) Optimization
  5. Advanced NLP Settings

Advanced NLP Settings

On the Kore.ai Bots Platform, natural language engine attempts to identify a user input and match that user input to a task. You can modify additional advanced settings like auto training setting for user utterances and fine-tune threshold levels for intent detection.

Navigating to the Advanced Settings

  1. Open the bot for which you want to add synonyms.
  2. Hover over the side navigation panel and then click Natural Language.
  3. Click the Advanced Settings tab.
  4. Below is a detailed discussion about each and every section on this page.

Auto-training Settings

Use this setting to initiate unsupervised learning, whereby the successful user utterances are automatically added to the ML training model.

Multiple Intent Detection

You can also allow the NLP engine to detect and execute multiple intents identified in a single user utterance. This setting is disabled by default and can be enabled. Once enabled, the NLP engine determines the sequence in which these intents are to be executed based on the standard phrases like ‘before that’, ‘after that’, ‘and then’, etc. in the user utterance. If no order is specified or identified by the NLP engine, then the intents are executed in the order they are present in the original utterance.

Threshold & Configurations

Threshold and configurations can be specified for all three NLP engines.


Fundamental Meaning

Proximity of Probable Matches defines the maximum difference to be allowed between top scoring and immediate next possible intents to consider them as equally important.


Knowledge Collection

Path Coverage can be used to define the minimum percentage of terms in the user’s utterance to be present in a path to qualify it for further scoring.
Definite Score for KG defines the minimum score for a Knowledge Collection or FAQ match to be considered as a definite match and discard any other intent matches found.
Minimum and Definitive Level for Knowledge Tasks allows you to set the confidence levels for a Knowledge Task. You can view and adjust the confidence level percentages for the task in one of three ranges as:

  • Not matching a task - The gray area represents the knowledge task NLP interpreter confidence levels are too low to match the task intent.
  • Multiple matches for a task - The orange area represents the NLP interpreter confidence levels range from minimum to maximum that provides one or more matches as a list of task intents to the user to choose from if the confidence level is more than 10% less than of the confidence level specified for Matching a task.
  • Matching a task - The blue-green area represents the NLP interpreter confidence levels range from minimum to maximum required to provide an exact match to one knowledge task intent provided no other task contains a score within 10%. If more than one task has a top matching score within 10%, the list of high scoring tasks is presented as a list of choices to the end-user.
To set confidence levels

  1. Click and drag the first slider level to the left or right, in 10% increments to define the minimum confidence level required to match a task.
  2. Click and drag the second slider level to the left or right in 10% increments to define the range for probable matches. If there is more than one probably match in this range, all probable matches are returned as a list of choices to the user. Any match with a confidence level higher than this setting is considered a match.

When the mouse drag is complete, changes are saved immediately and applied to the Bot.

KG Suggestions Count: Define the maximum number of KG / FAQ suggestions to be presented when a definite KG / FAQ match is not available.

Proximity of Suggested Matches: Define the maximum difference to be allowed between top scoring and immediate next suggested questions to consider them as equally important.


Machine Learning

For setting Machine Learning parameters, refer here.


Next Steps

To learn more about best practices and tips for optimizing NLP, see the Natural Language Processing Guide.


Was this article helpful to you? Yes 2 No