My First Repo - 2016 seattle home listing data analysis
The repository is the first repo I made on Github. I always aspired to become a future data scientist specialized in marketing, combining the quantitative analysis I learned in my Economic undergraduate education with the marketing knowledge I am learning now at Northwestern Univeristy.
This repo is a great opportunity for me to try as a data scientist, completing the Cross-Industry Standard Process for Data Mining (CRISP-DM) (Business Understanding, Data Understanding, Prepare Data, Data Modeling, Evaluate the Results, Deploy)
The dataset was introduced by Udacity - Data Scientest NanoDegree. It is a one-year Seattle Airbnb home listing data in 2016, which includes the listing price of the housing, and its housing information and review.
Three questions were defined and answered in our analysis
- What are the busiest times of the year to visit Seattle? By how much do prices spike?
- Can we use other listing information to predict the housing price?
- The vibe of each Seattle neighborhood using listing descriptions?
- Seattle_Airbnb_Analysis.ipynb :Notebook with detailed code and analysis of how these three questions are approached and tackled (latest update 10/14/2018)
- Summer (July, August) is the busiest season to visit Seattle. The price increased by 25% compared to the start of the year
- One lasso-regression was contructed to predicted the listing price with RMSE 3580.413, R² 0.571
- Five neighbourhood were selected and TF-IDF was calculated to identify the top 5 keyword for each
Thanks Udacity for providing this opportunity and the guidance of the mentor