Engineering path

UX architecture

Nowadays web and mobile interfaces are mostly built by information blocks, be it a post or a comment section with its usual interaction buttons. The aim of this section is to layout how information should be stored based on its own nature. We will dive into Ted Nelson's view together to how sentences are structured using IEML.

A gentile IEML intro

Between alike relations and oppositions, the main information substrate have one substance in common: Root concepts where everything else is created from. Having ontologies as the ground linking for the vast amount of information available is indispensable, hence why adopting the IEML language as a universal reference for a further semantic database.

Here we explore a mathematical ground with an algebraic notation to allow achieving computability of the previously proposed UX architecture.

Natural language processing

Simple grammatical classification can bring interesting results. Given a sentence, you will be able to know who's the subject and object, wether it as classified by an adjective and if it's performing an action through a verb changing the state of the object. Temporal and situational insights might also be inferred from adverbs and such. These classifications will work as a first step building sets of information under semantic networks.

Here we explore how the old school natural language processing together with recent AI innovations can help the generative and classification steps of such a system.