natural language processing

ARSINOE SustainGraph (Knowledge Graph)

Data Augmentation scripts for the parser MaChAmp-TWG as part of my bachelor thesis titled "Data Augmentation for TWG Parsing via Syntactically Well-formed Nonsense Sentences".

Slang Identification via Machine Learning using sklearn.

Slang Identification via Machine Learning using sklearn.

Generate ALT text (captions for low vision website users or book readers).

An attempt at AI-based lemmatization (or something close to lemmatization) of Old English

Master thesis - Deep learning-based pairwise alignment of protein sequences using transformer architecture.

Machine understanding surgical actions from intervention procedure textbooks

A repository containing assignment exercises of "Computational Argumentation" (SS21) course in Universität Paderborn

A small, RoBERTa-based language model for the Portuguese language.

Exploring and Visualizing Referring Expression Comprehension (Bachelor's Thesis by David Álvarez Rosa)

Python program for analysing text and inferring social relationships/interactions between people

Python application for automagically generating poetry.

This repository contains a running natural language processing with R. In the background, the huggingface (https://huggingface.co/models) model library is used (thus some of the code is written in Python). Beware this is work in progress.

System for assessing the strength of expression for selected phrases coming from natural language

The project uses natural language processing to automatically assess causal diagrams filled out by students based on semantic distance from the model causal diagram provided.

Trump claims to "have the best words", but a simple text analysis shows that a 16-year-old climate activist from Sweden has a larger English lexis than Trump. Also featuring Hannah Arendt, Simone de Beauvoir, and others.