In the recent years, smart home technologies, such as Amazon Alexa, have transformed the way users interact with their living environments. This study investigates sentiment analysis of customer reviews for Amazon Alexa products, including devices like Echo, Echo Dot, and Firestick. Leveraging machine learning techniques such as Logistic Regression, Convolutional Neural Networks (CNN), Artificial Neural Networks (ANN), and GloVe word embeddings, the objective of this research is to categorize reviews as either positive or negative. The motivation behind this study stems from the increasing influence of customer feedback on product development and user satisfaction. Understanding consumer sentiment helps businesses enhance their offerings, address user concerns, and improve smart home experiences. The evaluation is conducted using metrics such as recall, precision, and overall accuracy to determine model effectiveness. Results indicate that deep learning models, particularly CNN and ANN, outperform traditional approaches in sentiment classification. By providing insights into consumer sentiment, this research emphasizes the importance of robust sentiment prediction models for better decision-making in the smart home industry. Businesses can leverage these findings to refine product strategies, enhance user experiences, and drive innovation in the rapidly evolving smart home technology market.