A pipeline designed to detect political topics, hashtags, URLs, user mention, and hashtag-based voting intentions, expressed in tweets about the UK referendum on membership of...

Report on the Humane-AI microproject "Pluralistic Recommendation in News". It contains details on how the two relevant datasets were built, a link to the two, and details...

A machine learning classifier trained to categorise text about COVID-19 vaccines into 6 categories - Liberty or Freedom; Development, Provision and Access; Safety, Efficacy...

Base Noun Phrase Chunker, producing NounChunk annotations.

The CYMRIE named entity recognition is a service for the analysis of Welsh text. It identifies name of persons, locations, organizations, as well as money amounts, time and...

This is a named entity recognition service for documents in German. Based on ANNIE, it identifies names of persons, locations, and organizations.

Classifies short texts such as Wikipedia comments based on whether they contain toxic language

The SMAPH system links queries to the entities it mentions, disambiguating mentions if needed. Entities are Wikipedia pages. This problem is known as "entity recognition and...

An event classification package.

The SUMMA Text Summarization (EN) uses the SUMMA toolkit developed by Horacio Saggion to provide a generic English document summarizer.

English named entity recognition service. Identifies names of persons, locations, organizations, as well as money amounts, time and date expressions

This method recognises opinionated sentences in English text and it classifies them as positive or negative.It also indicates emotion type, author and target of the opinion,...

Attempts to classify whether a text is hyper-partisan or not. This system won first prize in the Hyperpartisan News Detection task at SemEval 2019

Service that takes two news article texts and attempts to determine whether they refer to the same underlying news story. The articles may be in the same or in different...

A service that tags abusive utterances in any text with the type of abuse (sexist, racist, etc.) and whether the abuse was aimed at the addressee or some other party. This can...

Service that processes threads of short texts such as tweets or forum postings and attempts to determine the stance (support, deny, question or comment) of each reply towards...

This method analyses French tweets for names of persons, locations and organizations. It also performs normalization of abbreviations and common Twitter slang.

This method is the English tokeniser, sentence splitter, POS tagger, phrase chunker and named-entity recogniser from Apache OpenNLP.

Named entity recognition service for Twitter data. Identifies person, location, organization etc. and also performs normalization of abbreviations and common shorthands (such as...