Here are some of the things I did, or I was involved in somehow, and software I maintain.
Unsupervised Topic-based Lexical Debias
Hurtlex is a multilingual lexicon of offensive, aggressive and hateful words. It was created starting from an original resource by linguist Tullio De Mauro and semi-automatically translated into many languages.
KNEWS (Knowledge Extraction With Semantics) is a software that bridges semantic parsing, word sense disambiguation, and entity linking to produce a unified, LOD-compliant abstract representation of meaning.
The collection of tweets from Twitter in Italian language.
HTTP RESTful API to analyze English natural language using the C&C tools and Boxer.
Delicious Folksonomy Dataset
A dataset obtained crawling Delicious, the social bookmarking website.
C&C/Boxer Web Interface
Web interface for the C&C/Boxer linguistic analysis pipeline.
Python module to search and download messages from Twitter.
GNU Octave implementation of the Listnet learning-to-rank algorithm.
The Groningen Meaning Bank
A large corpus of semantically annotated English text that anyone can edit.
A Game With A Purpose to collect linguistic annotation.
Word and sentence boundary detection software (a tokenizer, that is), based on supervised statistical methods.