paint-brush
Lets Study the Seattle Airbnb Databy@zaidzidane
119 reads

Lets Study the Seattle Airbnb Data

by ZaidMay 29th, 2020
Read on Terminal Reader
Read this story w/o Javascript
tldt arrow

Too Long; Didn't Read

The price of the property in Seattle, Seattle, is either occupied or remain empty. Being a superhost does not give any extra occupancy benefit. Being superhost is not an extra benefit for the owner. We can also explore on the effect of neighbhorhood on the bookings. We can use the data to create a machine learning model based on the rating on the price and some extra categorical and quantitative features. We will also use this data to build a machine-learning model.

Company Mentioned

Mention Thumbnail
featured image - Lets Study the Seattle Airbnb Data
Zaid HackerNoon profile picture

So, recently I started my Udacity Nanodegree on Data Scientist. To be honest the first project speaks about CRISP-DM which is CRoss-Industry Standard Process for Data Mining.Let's leave it apart and start working on what we learn from the dataset.

So three question came to my mind when I look at the data.

First.Do the price of the property changes over a year or do it remain constant?

Generally we see that during summer and winter breaks the price increases for the property. But to my suprise it remain constant.

I tried to plot as much graph but I just saw that there was yes and no fluctuations.That is property is either occupied or remain empty.This also shows that the AirBnB does not support dynamic pricing which is actually cool.

Second question that comes to mind is whether low cost property is occupied more time than the rest or the duration of occupancy is not related to pricing.

So the property with the maximum price cost around 1650 and when I compared it occupancy it shows mere 60%.

Third question to mind is being a superhost matters and did it make extra bucks for the owner.

Here I had an interesting finding. I found out that being superhost does not give any extra occupancy benefit.

Rather the results came out that people that are not superhost have 86% occupancy whearas people who are superhost have just 83 % occupancy.

In Conclusion

With this dataset we can also explore on the effect of neighbhorhood on the bookings.

Location and their property charges can also be correlated.

We can also buld a machine learning model based on the rating on the price and some extra categorical and quantitative features.