← Back to Projects

Real Estate Data Analysis/Data Mining

Technology: Python, Pandas, Matplotlib, Seaborn Date: May 2026

Overview

This project was a start to finish implementation of the CRISP-DM methodology, using a sample real estate dataset. The dataset was provided in an excel file, leaving me to clean and prepare the data for analysis. The project was then implemented using the CRISP-DM methodology, and the results were analyzed and reported on.

Objectives

  • Practice with the CRISP-DM methodology
  • Practice with the Python data analysis toolkit
  • Become familiar with critical libraries like pandas, matplotlib, and seaborn

Methodology

The project started with a messy dataset. Then following CRISP-DM as a framework, we were briefed on the business questions involved (should this real estate company expand into short term rentals in Nashville?) and context behind the project. I then cleaned the data preparatory to exploratory analysis. Once the data was undestood, we prepared it for modeling and built models to determine what kinds of properties would make the best investments.

Key Findings / Results

Based on the data, Nashville was a good market to invest in. Additionally, larger properties (based on how many people they can accommodate) seemed to be the best investments.

Code Samples

Below are some code samples from this project:

Sample 1: Transforming variables for correlation analysis

df_corrT = pd.DataFrame()
# create new SQUARE ROOT transformed variables
df_corrT['calculated_host_listings_count_sqrt'] = np.sqrt(df_corr['calculated_host_listings_count'])
df_corrT['calculated_host_listings_count_sqrt']=np.sqrt(df_corr['host_listings_count'])
# run a spearman correlation analysis on your dataframe, then store values in cor_matrix
cor_matrix = df_corrT.corr(method = 'spearman')
# display the formatted matrix (i.e., with background gradient, etc.)
cor_matrix.style.background_gradient(cmap = 'seismic', axis = None, vmin = -1, vmax = 1)

Technologies Used

  • Python
  • Pandas
  • Matplotlib
  • Seaborn

Lessons Learned

This was a great way to flex some of my coding muscles and apply the skills I have learned. I also appreciated the opportunity to practice within an industry standard framework like CRISP-DM.