For my BAS 474 final project, I analyzed Airbnb listings in Austin to uncover meaningful insights for both Airbnb and its users. The report is split into two main parts: a cluster analysis to segment listings into distinct groups, and a predictive modeling section to estimate listing prices based on various features. In the cluster analysis, I used K-means to group similar listings and interpreted each cluster based on factors like room type, number of reviews, and price range. This could help Airbnb identify different types of listings on their platform or give travelers better ways to filter options. The second half focuses on price prediction. I tested several machine learning models and selected the one that offered the best balance of accuracy and interpretability. The final model can predict listing prices within roughly $1000 of the actual price on average. I also highlighted a few key factors, like number of bedrooms and host experience, that influence pricing, and suggested ways to improve accuracy with more data. The goal was to keep the report concise, visually clear, and useful to a non-technical audience, while still reflecting the technical decisions behind the analysis.

This analysis applies hierarchical clustering and k-means clustering to the SURVEY10 dataset using RStudio to identify distinct student segments. The goal is to develop data-driven marketing strategies for a new dating app based on behavioral and demographic insights. Using dplyr, tidyverse, and ggplot2, we preprocess the data, determine optimal clusters, and visualize segment characteristics. The final report presents key findings, segmentation rationale, and two targeted marketing strategies tailored to the identified clusters. The results provide actionable insights to enhance user engagement and app adoption among different student groups.