Offspring from Reproduction Problems (invitation to discussion and announcement ACL presentation)

Despite clear methodological descriptions and making our code and data available, it is still difficult to make our research reproducible. We would like to invite the NLP community to join a discussion on reproducing results from NLP research. One of the main questions we want to address is how we can get a better understanding of our methods through systematic research. Please visit our website: to join our discussion group or share your experiences and ideas. …or come and talk to…

Read More

Antske Fokkens to present at Annual Network Institute Event

Antske Fokkens will present at the Annual Network Institute Event on July 4. During this event, the different groups working together in the interdisciplinary Network Institute at VU University will come together to exchange their research ideas and results. Antske will describe her attempts at reproducing some research published by others in her field, using the original data. And guess how easy that was…  

Read More

NERD-ML 2nd in MSM2013 Concept Extraction Challenge

Together with Giuseppe Rizzo and Raphaël Troncy from EURECOM, NewsReader team member Marieke van Erp participated in the MSM2013 Concept Extraction Challenge with a system called NERD-ML. This system was developed  to identify and classify named entities in microposts (Tweets). Their system ranked 2nd in the challenge, only 0.01 points in F-score behind the best system. NERD-ML is built on top of the NERD framework developed at EURECOM that makes it possible to easily query different named entity extractors that are available on the web. With NERD-ML, this framework…

Read More