Identifying Annual Consumer Homestay Activity of Top-Rated Airbnbs through Simulation and RNG Modelling

Airbnb has been quite the competitor in the hotel industry alongside accommodating users as they continue to travel around the world. Looking back to how it started back in 2008 in San Francisco, California, United States, its surreal to imagine how the community had built upon itself to become self-sufficient enough to allow its reputation system be the very foundation of credibility among those who are guests and those who are hosts. This study goes in-depth on which Airbnb rentals offer the top-rated accommodations, as factors involved are taken into account. These factors are used in a random-number model to take the substantial dataset of those Airbnb rentals in New York City as machine learning techniques are used to employ predictive modeling identifying property desirability and occupancy rates. This simulation model starts off computationally intensive as randomly generated numbers are assigned to each property per day, and is optimized through parallel processing to enhance the performance and efficiency of the model. The small differences made in the configurations in optimizing the unoptimized model have then efficiently generated a noticeable difference in the runtime. This is then concluded as the optimizations made in the data-driven approach used in the model can be utilized by property hosts and other industry practitioners in making decisions for the homestay market.