Research Project

NLX : Research Project with Model Based NLP

NLX : Research Project with Model Based NLP

Share this
Tweet this
Email this

NLX aims to provide a grammar trainer to derive a formal but trainable model by supervised learning methods to resolve the grammar tree from natural language text and creates a knowledge graph from it. The project originated before the rise of GraphRAGS and LLMs which now do the job quite reliably and made this project obsolete right now. But on the long term we are still convinced, that natural language grammar can be resolved better with a formal model than a statistical one like LLMs are. The reason is because the fundament of natural language is a deterministic grammar

Follow @XIXUM Sponsor Watch Star
Product Features

  • Natural Language DSL

    A designed Domain Specific Language (DSL) is used to tokenize natural language text and interprets it semantically to convert it on to a Document Object Model (DOM)

  • Grammar Trainer

    A spatial Graph Model is used to store the training data in a formal shape of relational semantics and thereby detect the decision criteria for the grammar structure on various abstraction levels of the grammar tree. Grammar logic is decided on certain abstraction level thus simple formal rules on singular and atomic values is not enough. It's a deterministic pattern that resolves on higher meta layers and abstraction levels

Product Overview
Share this
Tweet this
Email this
Product Overview

Sed augue ipsum, egestas nec, vestibulum et, malesuada adipiscing, dui. In auctor lobortis lacus. Praesent ac sem eget est egestas volutpat. Sed magna purus, fermentum eu, tincidunt eu, varius ut, felis.

URLs

We are a group of skilled developers and programmers.

We Are A Group Of Skilled Programmers

Besides our awesome product palette we offer a number of additional service. Click the navigation button below to discover more about our service portfolio