Machine learning vs. deep learning for Starbucks sentiment analysis

A new study compares machine learning and deep learning models for analyzing customer sentiment in the retail coffee sector, specifically focusing on Starbucks reviews from ConsumerAffairs.com. The research evaluated five traditional machine learning models and five deep learning models, finding that Support Vector Machine (SVM) achieved 91.0% accuracy among the former, while Bidirectional LSTM performed best among the deep learning approaches. The study also noted that class imbalance in the dataset negatively impacted the recall for positive sentiment across multiple models, underscoring the importance of model selection and preprocessing for customer experience analytics. AI

IMPACT Highlights the effectiveness of specific deep learning models like Bidirectional LSTM for real-world sentiment analysis in customer service.

RANK_REASON Academic paper detailing comparative analysis of ML and DL models for sentiment analysis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• CNN
• ConsumerAffairs.com
• convolutional neural network
• decision tree
• gated recurrent unit
• logistic regression model
• long short-term memory
• Md. Alamgir Hossain
• naive Bayes classifier
• random forest
• recurrent neural network
• Starbucks
• support vector machine

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →