Multi-purpose framework for automated text understanding
TextSpace is a flexible framework which helps developers to build Natural Language Understanding (NLU) and Natural Language Processing (NLP) solutions, with intent classification and entity extraction as its two major components. The solution can be employed on the clients' own machines, hence do not require to be employed in the cloud where data privacy and GDPR issues arise. Furthermore, TextSpace comes pre-trained on 36 languages, including languages with small data sets like the Scandinavian and Indian languages.
The main features of TextSpace include:
- Intent Classification: Parse the input text to detect the intentions behind the written words.
- Named Entity Recognition: Identifies entities in text. Classifies these into a set of standard and custom categories.
- Summarization: Takes a document as input and outputs a summary with its most important content and main ideas.
- Neural Machine Translation: Translations customized for specific purposes (e.g. insurance policies or legal texts) and available in rare languages.
Currently, we work on expanding the capabilities of TextSpace to include a cognitive answering machine, a data enrichment tool, a spell checker, a sentiment analyzer, a language recognizer, speech to text, and topic modeling.
Please get in touch for further information or a demo: