AIRBNB FINAL PRESENTATION GUIDELINES Presentation Structure For your final presentation, as specified in the Airbnb Project Overview Guideline, you are expected to integrate all three components: 

 • Part 1: Problem Identification and Research Question Formulation o Define the central issue and frame your investigative questions. 

 • Part 2: Data-Driven Analysis for Insight Discovery Utilize a range of analytical approaches to address your research questions: o 

Unsupervised Learning: 

 ▪ PCA/Factor Analysis to reduce dimensions and identify patterns.

 ▪ Clustering to categorize data points into meaningful groups. o Supervised Learning: utilize Classification Models to predict outcomes. o Text Mining:

 ▪ Analyze Word Count to quantify word usage.

 ▪ Conduct Sentiment Analysis to assess the tone within the text. 

 ▪ Perform Topic Modeling to uncover underlying themes. 

 • Part 3: Recommendations Based on Analytical Insights o Present actionable strategies derived from your data analysis. Contrary to the midterm presentation, for your final presentation, you have the flexibility to pursue various questions within each analytical method in Part 2 due to the exploratory nature of these analyses. The midterm presentation required a sequential analysis that led to a unified set of recommendations, but for the final, you may develop specific, individualized recommendations for each separate question you investigate. For your final presentation, the sequence of content is flexible and need not strictly adhere to the prescribed structure (Part 1, then Part 2, followed by Part 3). Instead, you may opt to present each analytical method independently by first posing a research question, then sharing the corresponding analysis and insights, and concluding with your recommendations before proceeding to the next method and repeating the process.

Identifying a Problem For the final project, you'll again take on a specific role within the Airbnb platform. You have the option to maintain the role you assumed for the midterm or select a different one. Your choices are limited to: a host, a guest, or a representative of Airbnb itself. Please remind us of the city you're focusing on and which dataset(s) you're exploring. If you decide to concentrate on one or two neighborhood groups within a city, describe the data cleaning process and justify this approach. Define the problem you intend to investigate. What is your research question? What aspects would you like to explore? You can formulate one overarching question that can be addressed using all analyses in Part 2. Alternatively, you may choose to pose different questions (while retaining the same role) for each analysis in Part 2. Unsupervised Learning PCA/Factor Analysis For your PCA/factor analysis, start by selecting continuous x-variables, excluding the dependent variable (Y-variable) used in your midterm presentation (though you can choose a new one for this analysis). Include as many x-variables as possible to ensure robustness. Analyze and describe the following:

 ▪ The number of factors extracted from the analysis. 

 ▪ The percentage of variance in the original data captured by these factors.

 ▪ Label the first three factors clearly. If you extract fewer than three, label each one. Next, include these factors as predictors in a linear regression model predicting your selected Y-variable. You may additionally include nominal variables in the regression that were not initially considered in your factor analysis. Evaluate and explain the following: 

 ▪ The most influential factor in your model. If it's a new factor not labeled earlier, assign an appropriate label.

 ▪ The overall performance of the model. If you've used the same Y-variable as in the midterm, compare this model's performance with the linear regression from your midterm presentation