Automating Sentiment Insights with VOC AI for Improved Customer Experience and Operational Agility For a TV Home Shopping and Retail Platform

Solution

• Automated Sentiment Classification: Built using Python and HuggingFace for NLP to automatically classify customer feedback into Positive, Negative, or Neutral sentiment.
• Experience and Department Mapping: Implemented with custom Python logic and Power BI to associate sentiment with specific customer experience categories.
• GPT-powered VOC-Assist Interface: A conversational interface built with LangChain and Streamlit that allows stakeholders to query sentiment insights using natural language.
• Anomaly Detection for Proactive Alert: Python-based analytics module that performs quarterly anomaly detection in customer feedback trends.
• Interactive Dashboards for Insights: Developed in Power BI to display insights by Product, Customer, Brand, Vendor, Department, Experience, and Time.

Outcome