A team of Indiana University students (Elizabeth Supinski, Hui Chen, Rusty Hann and Ding Li) explored the relationship of textual and visual emotion by creating a fascinating analytics pipeline involving a number of tools. Their paper includes the core workflow and a number of code samples capturing how they analyzed the data, as well as commentary on the unexpected complexity of analyzing data at these scales, offering a good starting point for others interested in building on their work and they even created an interactive demonstration site showcasing the final results!
IU: Visualizing Our Global World: Correlation Between Article Tone and Emotion of Accompanying Images
