The HUES (Heartfelt Utterances Emotive Strokes) project came from a simple question: Can art help us understand our feelings better? We all know that art can inspire many emotions in us. But what if we could use art to explore and express our emotions in the digital world? Social media has become a significant way for people to share their lives and emotions. Users post their joys, sadness, and everything in between. The project also draws inspiration from how people use digital tools like Photoshop to edit and create images. While these tools are based on complex code, HUES takes a different approach. Instead of users manually editing images, HUES automatically applies artistic effects to social media content. This offers a new way to view and understand emotional expression in digital spaces. To do this, the project utilized two datasets from Kaggle: the Emotion Recognition Data and the Twitter Emotion Dataset. These datasets provide a good mix of images and text.

The process began with data cleaning and file renaming to improve usability. During data preprocessing, I went through all the sentiment texts and created word cloud visualizations to see which words appeared most frequently in each sentiment. In the word cloud visualization with the anger messages, the most common words include "feeling," "people," "time," "know," and "want." These words appear frequently because they reflect common themes of frustration with others, feelings of being wronged, and desire for change in emotional charged situations. In the word cloud visualization related to happiness and surprise messages, the most common words include "feel," "people," "life," "love," and "time.” These words are commonly used to express joy, appreciation, and positive emotions, often reflecting their relationships, experiences, and moments of fulfillment or surprise in their lives. Common words in the sadness-related word cloud visualization include "feeling," "people," "time," "know," and "life." These words often appear when people express their emotional state, often reflecting relationships, personal struggles, and moments of isolation or regret. The emphasis on “feeling” and “living” suggests introspection about one's emotional well-being and life circumstances in moments of grief.

The final dataset includes 10 images and about 20 messages for each emotion. OpenCV, a computer vision library, was used to edit each image. Five different artistic styles were developed: drawing, surrealism, watercolor, poly, and halftone. Each time the program runs, it selects a random style to apply to each image. This approach keeps the output fresh and interesting. The images don't appear all at once. Each one displays for about 10 seconds, with potential delays based on the complexity of creating the new style. While the images change, related messages float across the screen. This creates a digital art show that combines visuals and words to express different emotionsThe images don't appear all at once. Each one displays for about 10 seconds, with potential delays based on the complexity of creating the new style. While the images change, related messages float across the screen. This creates a digital art show that combines visuals and words to express different emotions.

Below are 11 images representing four emotions (anger, happiness, sadness, and surprise), each with an applied style. You can click on any image to view the full sentence of each floating message.

HUES invites people to consider how feelings are shared in the digital age. When an image is transformed into a different style of art, it can change people’s emotional response to it. A photo that initially evokes a sense of happiness may suddenly seem mysterious when given a surreal effect. A sad post may feel more hopeful when turned into a watercolor. Floating messages add another layer of interpretation. They remind people that words and images can work together to express emotion. Sometimes they match perfectly, while other times they seem to contradict each other, reflecting the complexity of real emotions.

By expressing emotions through art and texts, HUES encourages audiences to stop and think about what they see and feel. HUES is also about connection. It utilizes technology and art to bridge the gap between inner feelings and outer expression. The project asks audiences to look beyond the surface of social media posts and consider the real emotions behind them. By transforming familiar social media content into art, it offers new perspectives on how people communicate their feelings online. This artistic approach to data visualization offers a unique way to explore the intersection of technology, art and human emotion in the digital age.