Ongoing Projects

Californians Talk HIV: Using Social Media to Get Steps Ahead

Multidisciplinary project with public health and computer science researchers from the University of California San Diego and the Johns Hopkins University, aiming to develop methods to extract actionable intelligence from social media platforms to support timely, evidence-based decision making and help public health leaders and advocates stay one step ahead of California’s HIV epidemic.

drawing

Increasing Citizen Participation in E-Rulemaking by Harnessing Social Media

Multidisciplinary project with public health and computer science researchers from University of California San Diego and the Johns Hopkins University, aiming to increase public participation in the regulatory process by crowdsourcing and extracting actionable feedback regarding government regulations from public social media comments.

drawing

Past Projects

EXPRESS

Expression and Recognition of Irony in Multicultural Social Media

Joint research project funded by FCT Portugal with researchers from INESC-ID Lisboa and the University of Texas at Austin, on the automatic analysis of sarcasm and irony in social media.

We developed novel sarcasm detection methods, including CUE-CNN — a deep neural network that automatically learns and exploits representations for both the author and contents of a post to make contextualized inferences over highly ambiguous content.

drawing

drawing

Press Coverage

New Scientist Digital Trends Tech Crunch Sky News Daily Mail Sapo
Tech Xplore Improbable Research Evening Standard Euclid Vocativ  

POPSTAR drawing

Public Opinion and Sentiment Tracking, Analysis, and Research

Interdisciplinary R&D project funded by FCT Portugal with researchers from with INESC-ID Lisboa, Instituto de Ciências Sociais, Universidade do Minho, Faculdade de Engenharia da Universidade do Porto, on using conventional and social media to measure and track public opinion about political entities.

We built a political opinion mining for social media to measure the popularity and attitudes about Portuguese political entities over time. The indicators produced by the system were aligned with traditional polls and published in quasi-realtime (i.e. updated daily) in a publicly available dashboard, which has been used by political scientists to conduct new studies on public opinion.

drawing

drawing

drawing

See the project’s (archived) webpage for more details and the respective scoop.it for some of the related press coverage.

drawing

REACTION drawing

Retrieval, Extraction and Aggregation Computing Technology for Integrating and Organizing News

Interdisciplinary R&D project, funded by the CoLab, UT Austin - Portugal, involving researchers from the Faculdade de Ciências da Universidade de Lisboa, INESC-ID Lisboa, Faculdade de Engenharia da Universidade do Porto, Centro de Investigação Media e Jornalismo, University of Texas at Austin and news media partners such as Sapo and Público to create tools and methodologies to support advanced computational newsrooms.

We developed novel data-driven tools for journalists to gain new insights into complex topics — e.g. our tools were used in an article about political campaign financing that won a cyber journalism award. We also developed various tools for journalists to use social media to track media personalities over time, gauge public reactions to newsworthy events, and analyse global events in real-time.

drawing

drawing

See the project’s (archived) webpage for more details and the respective scoop.it for some of the related press coverage.

drawing