Collaborative Research: HNDS-I: NewsScribe - Extending and Enhancing the Media Cloud Searchable Global Online News Archive

University Of Massachusetts Amherst, Amherst MA

Investigators

Abstract

The news that people read online forms how they understand the world around them. To study the relationships between what news is reported and how that affects human behavior, researchers need an easy way to search through lots of news articles. NewsScribe is a tool that lets scientists, educators, students, and the public search through over 1.5 billion online news stories from around the world. It adds about 500,000 new stories every day to stay timely and relevant. NewsScribe makes it easy to find information without needing to create or seek out large volumes of text. It organizes news sources in different collections by location and topic. This organization helps researchers quickly find the stories that are relevant to the questions they are studying. The activities supported by this award include updating the software and data at the core of NewsScribe and adding new user-friendly features and visualizations to make the tool easier to use. NewsScribe makes it possible for different types of researchers to study patterns in news reporting from around the world in new ways, at broader scales, over the last few years or the last few days. There are many challenges involved in creating and maintaining such a large database while ensuring it will still be useful for so many different users. This project develops software that quickly collects and stores online news articles by processing them to pull out information like the publication date, language, and just the content of the article. New approaches to that process allow for pulling out more accurate data, handling more stories, and also returning search results faster. Various levels of alert systems notify the software team when some problem occurs with the daily processing of news stories, ensuring that the system is stable and reliable. The web-based search tools showing these results can sometimes hard to use, with a mix of tables of information, charts and graphs, and downloadable reports. Testing with current and new users guides continuing development of more user-friendly approaches, stabilizing the NewsScribe product and ensuring its long-term usefulness. The award supports not only the development of improved software but also establishing an advisory board of close partners who help guide how NewsScribe can best support the network of people who rely on it. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.