Improving Airbnb Customer Satisfaction with Text Mining & Sentiment Analysis

What I Did:

• Analyzed 354+ customer reviews using text mining (IBM SPSS Modeler) to uncover key drivers of satisfaction.
• Categorized feedback into 8 themes (Host, Cleanliness, Amenities, etc.) and classified sentiments (Positive/Negative).
• Applied CRISP-DM framework to clean, model, and validate insights from unstructured text.

Key Results:

• Top Pain Points: 37% of complaints targeted host attitudes, while 35% cited cleanliness issues – direct opportunities for improvement.
• Strengths: Guests loved neighborhood vibes (22% positive) and accessibility (17% positive) – areas to promote.
• Actionable Insight: Proposed host training programs and stricter cleanliness checks to address critical gaps.

Tools: IBM SPSS Modeler, Excel | Skills: Text Mining, Sentiment Analysis, CRISP-DM, Data Cleaning