Research on Social Interaction and Music Appreciation Behavior in IoT Music Platforms

Music platforms are increasingly incorporating smart devices to provide personalized and interactive music experiences with the development of Internet of Things (IoT) technology. These platforms give consumers new methods to connect with music, which influences their social relationships and music enjoyment habits. However, the connection between IoT-driven features and user behavior in music consumption is primarily undefined. The purpose of this research is to establish the social interaction and music appreciation behavior in IoT music platforms. It utilized a dataset consisting of 168 participants who regularly interact with a popular IoT music platform, capturing qualitative feedback through user surveys. The data were preprocessed using Z-score normalization and a bag of words (BoW) from the acquired data. The TF-IDF is used to extract the features. It presents a novel Advanced Northern Goshawk mutated Gradient Boosting Machine (ANG-GBM) to predict the music appreciation behavior frequency on social interaction in music. ANG can be used to optimize the hyperparameters of the GBM model, ensuring that the final model achieves high accuracy in predicting music appreciation behavior based on social interactions. The suggested method is implemented with Python software, and its comparison is also given with the traditional algorithms. It achieves an F1-score of 91.7%, precision of 92.0%, recall of 91.5% and accuracy of 93.2%. The result shows that the proposed method outperforms existing methods in high accuracy of music appreciation behavior frequency. Finally, it highlights the potential of IoT and machine learning to make music listening more dynamic and engaging.