A Framework for Measuring Customer Satisfaction and Product Recommendation for Ecommerce

Tech Stack: Python , NLTK, Tableau, Weka, Hadoop

• Pre-processed amazon product reviews of phones and cameras using NLTK and identified frequent features.
• Parsed the unstructured reviews and opinion words about the extracted features with the help of Stanford Parser.
• Assigned scores to features using SentiWordNet and classified sentiments as positive/negative/neutral.
• Visualized results in an intuitive and interactive fashion using Tableau.
• Recommended top products to customers on the basis of performance of their features.