Natural Language Understanding
Find the meaning from unstructured text using Ionkom machine learning.
Intelligent text analysis
Natural Language Understanding (NLU) enables computers to understand human language contained in unstructured data and deliver critical insights.
NLU uses machine learning to predict text insights like user intention and domain specific entities. You can extract information about people, places, and events, and use them to better understand customer conversations.
Chatbots and customer support
Combining Language Understanding and Dialog System - Ionkom AI enable customers to create conversational interfaces for various scenarios like banking, travel and entertainment. In cloud natural language models always learn, you can benefit from the latest versions using provided APIs.
Enterprise-ready, available on premise
The service is ready to be deployed in existing commercial applications and can scale with enterprise quality and performance. The service meets international compliance standards, supports unlimited API calls, making it highly accessible in your organization.
Custom language solution
Quickly build a custom language solution, just bring your data, no coding skills required. With our support, the models can benefit from always learning. Active learning support is used to continuously improve the quality of the natural language models.
Two classifiers one model
Underlying ML Technology - The algorithm uses state of the art natural language understanding approaches. It uses a sequence to sequence neural network, that encodes the meaning of a sentence and two decoders, one for intent detection and another for slot filling (or entity recognition) that decodes the captured meaning.
Intent detection: Classify text by sentence meaning. Find out what your customers want, in the example below 'flight' will be the predicted intention.
Entity Recognition: Extract relevant information from unstructured text. Entities can be places, dates, persons in a general context or more precise like in the example below entities dependent on sentence context like 'from location' and 'to location'.