FD (Financieel Dagblad) Internships and MSc projects
Internships at FD (Financieel Dagblad)
The FD Mediagroep (FDMG) is the primary source of financial economic news in the Netherlands, through the daily financial newspaper “Het Financieele Dagblad” (FD), and the only commercial news radio station in the Netherlands (“BNR Nieuwsradio”). Serving a wide variety of users with different backgrounds, interests, and contexts, we believe that serving our digital media (text and audio) in a personalized manner can be beneficial.
Currently we are looking for interns for various project related to the personalization of news experience. Working at FDMG means that you get to work with all the news articles published by FD since 1985 or with the audio recorded by BNR since 2002 as well as our userdata. Also, we focus on making actual products, so it is likely that your code will be used in future versions of the FD app or BNR app.
Below is a list of ideas for internships for master-level interns (projects at least 3 month long). If you are interested, contact Arjen with a brief explanation what attracts you in the project and why you qualify, and he will introduce you to Maya Sapelli from FD.
Project 1: Categorization of news articles of FD (Financieel Dagblad)
FD has a large collection of news articles which are currently labeled with categories (multilabel,multiclass) by the journalists themselves. In order to support the journalists, we want to be able to suggest them possible categories, or label the articles automatically. In this project you will try out several algorithms, such as SVM vs logistic regression, or methods, such as hierarchical classification or flat classification for automatically categorizing news articles into categories in order to improve upon the current tagger.
Project 2: Automatic Summarization of news articles of FD (Financieel Dagblad)
At FD we are currently working on developing methods for automatically summarizing news articles. In this project you will implement a summarization pipeline from literature and compare the results to the system which is already in place. Depending on your preference you can choose for an NLP based approach to summarization, a Deep Learning based approach or a hybrid approach. FD articles and corresponding summaries are available as training material.