Exploring Sentiment Analysis in Digital Design: a Machine Learning Approach to Analyzing User Perceptions of Canva

This study investigates user sentiment towards Canva, a leading graphic design platform widely used by individuals and businesses for creating visual content. By analyzing over <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{5, 0 0 0}$</tex> user reviews collected from the Google Play Store, the research applies advanced machine learning techniques to classify sentiments expressed in user feedback. Three popular classification algorithms-Support Vector Machine (SVM), Random Forest, and K-Nearest Neighbors (KNN)-were evaluated to determine their effectiveness in sentiment analysis tasks. The data preprocessing pipeline involved rigorous cleaning, normalization, and stemming to ensure high-quality input for the models. Among the algorithms tested, SVM demonstrated the highest accuracy at <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 4. 2 \%}$</tex>, closely followed by Random Forest at <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 3. 8 \%}$</tex>, and KNN at <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 1. 9 \%}$</tex>. The sentiment analysis results reveal a predominantly positive user perception, with frequent praise for Canva's user-friendly interface and extensive free features. However, some negative sentiments were identified, primarily related to dissatisfaction with subscription pricing. These insights provide valuable feedback for Canva's product development and marketing strategies, while also contributing to the broader field of sentiment analysis by showcasing the comparative performance of different machine learning classifiers on real-world app review data.