Scope Analyzed a dataset of clothing reviews to uncover customer sentiment and engagement trends. Combined text data, profile sources, and review frequency to create actionable insights for retail and e-commerce teams.

Preprocessed review text using Python (NLTK): tokenization, stopword removal, and lemmatization.

Applied sentiment scoring to classify reviews as positive, neutral, or negative.

Combined sentiment results with review frequency to build customer engagement categories.

Designed a classification rule to flag highly engaged customers leaving multiple negative reviews (“alert” category).

Visualized results in interactive Power BI dashboards for quick decision-making.

Demonstrated how data-driven sentiment analysis can bridge technical analytics and business action.

Showed how retail teams could proactively address concerns, reduce returns, and improve retention.

Produced a clear, visual example of how insights can be shared with both technical and non-technical audiences.

Python (NLTK)

Power BI

SQL

Sentiment Analysis

Data Visualization