In this hands-on session, Melvin Wevers will show how he extracted a particular type of articles, namely recipes, from digitized newspapers. Using machine learning and an existing collection of annotated modern recipes, he managed to extract over 27 thousand recipes from four newspapers between 1950 and 1990. Also, he will showcase how nutritional information can be extracted from these recipes and how they can be automatically classified using tags.
Melvin Wevers is a postdoc in the DHLab at the KNAW Humanities Cluster in Amsterdam. His research interests include the study of cultural-historical phenomena using computational means with a specific interest in the formation and evolution of ideas and concepts in public discourse. In addition, he really likes to study advertisements and consumer products. Melvin has a background in American Studies, Psychology, and Cultural Analysis.