Sentiment Analysis and Satisfaction Evaluation of Smart and Connected Products Considering Interpretability

Smart and connected products (SCPs) integrate IoT and artificial intelligence technologies, relying on embedded sensors and network connections to achieve real-time interaction, and have become the core driving force for the development of smart home, Industry 4.0 and smart cities. As an important data source that reflects consumer needs and satisfaction, user reviews provide a key basis for optimizing product design and improving user experience. Aiming at the shortcomings of existing sentiment analysis methods in complex semantic modeling, fuzzy linguistic quantification and interpretability, this paper proposes a comprehensive sentiment analysis framework and satisfaction analysis method. Firstly, a bidirectional long short-term memory (BiLSTM) technique is used to accurately capture the semantic association of commentary text. Secondly, the probabilistic linguistic term set (PLTS) is utilized to quantify the uncertainty of fuzzy linguistics. Furthermore, the contribution of features to emotion prediction is revealed through Shapley additive explanations (SHAP). Finally, SHAP and importance-performance analysis (IPA) are combined to generate optimized priority recommendations. Taking smart speakers as an example, the effectiveness of the proposed sentiment analysis framework is verified by experiments. The F1 score of the test set is 0.967 and AUC is 0.979, which is significantly better than traditional methods. Therefore, this paper not only helps to improve the sentiment analysis of SCPs, but also provides scientific basis and practical guidance for product function optimization and personalized recommendations.