Study Predicts Ad Preferences Using Physiological Monitoring
A groundbreaking study has been conducted to predict consumer preferences in video advertising using physiological monitoring tools. The research, carried out by an unspecified university, employed Electrodermal Activity (EDA) and Facial Expression Analysis (FEA) to gauge reactions to ads.
The study utilised an eXplainable AI module to pinpoint the most influential features. These were identified as Attention, Engagement, Joy, and Disgust. The research suggests that these systems can accurately predict consumer ad preferences based on videos.
An AI system was developed using machine learning techniques such as k-Nearest Neighbors, Support Vector Machine, and Random Forest. Among these, Random Forest (RF) emerged as the top performer with an impressive 81% accuracy, 84% precision, 79% recall, and an F1-score of 81%. The study also found that emotions like Joy, Disgust, and Surprise played a significant role in preference prediction.
In conclusion, the research demonstrates the potential of physiological monitoring tools in predicting consumer ad preferences. The use of Random Forest in the AI system showed promising results, with high accuracy and precision. Further studies could build on these findings to refine and enhance the predictive capabilities of such systems.
