Intent detection in Kore.ai Bots Platform


(Madhu) #1

The Kore.ai Bots Platform includes a multi-pronged approach to natural language, which powers the way chatbots communicate, understand, and respond to user requests – and it is automatically enabled for every pre-built Kore.ai bot you use and the bots you custom build on the platform. Our hybrid NLP strategy was developed internally (not from another vendor’s services) for optimal outcomes and includes:

  • A computational linguistics based approach called Fundamental Meaning (FM) that’s built upon ChatScript. The model analyzes the structure of a user’s utterance to identify each word by meaning, position, conjugation, capitalization, plurality, and other factors.

  • A custom Machine Learning (ML) based approach. The Platform’s ML-based approach uses state of the art NLP algorithms and models.

The Platform combines these two distinct NLP approaches to enable you to instantly build conversational bots that are useful for up to 70% of conversation - with no language training to get started.

In addition to detecting and performing tasks (changes to system of records), Kore.ai provides an ability to build bots that can respond to frequently asked questions that return static responses. The platform uses the power of knowledge graph based model that provides the intelligence required to represent the importance of key domain terms and their relationships in identifying user’s intent (in this case the most appropriate question). Machine learning models append the Knowledge graph to further arrive at the right Knowledge query.

Once all the engines return scores and recommendations, Kore.ai has a ‘Ranking and Resolver’ engine that determines the winning intent based on the user utterance.