There are a lot of steps involved in dealing with any dataset. Some of them are listed below:
- Importing the libraries
- Loading and Cleaning the dataset
- Doing the exploratory data analysis (EDA) over the dataset, to gain some insights about the features and data
- Use machine learning models and figure out which one is the best/ or could be useful in making future predictions
One can look at the dataset differently and based on my understanding below are some of the questions that I will be discussing here:
- What is the busiest time of the year to visit Seattle? By how much do prices spike?
- Can you describe the vibe of each Seattle neighborhood using a listing description?
- Which factors do the customers look at while looking for a place?
Let’s roll now and discuss what we have gained from the dataset
……………………………………………………………………………………….
- What is the busiest time of the year to visit Seattle? By how much do prices spike?
PART-1: Let’s find out which is the busiest time of the year?
On the basis of the dataset, below are the Number of listings per month for the year 2016
In the above we can see that the busiest time of the year in Seattle is March means summer and we can see some increasing trend in December means winter, so it tells us that summer is the busiest followed by winter.
PART-2: Let’s see by how much the prices spike
Below is the Average Price variation per month.
Here we can notice that prices are more in months of June-July-aug-sep, it may be due to the fact that there are fewer listings available when we compare it with the above graph
- Can you describe the vibe of each Seattle neighborhood using a listing description?
In this question what we are trying to understand is how the location price varies monthly. Below is the plot of what we have achieved so far
From this one thing, we can notice is that in the month of March where we have most visitors people can stay in areas like University District or lake City as these areas offer cheap rates. Also, we can see that places near University District offer very cheap rates, whereas the areas like Magnolia, Cascade, Northgate offer expensive deals throughout the year.
- Which factors do the customers look at while looking for a place?
On using the best machine learning model, below are the best features that people look for while looking for a place.
From the above, we can see that some of the best features are: Bathrooms, how many people they can accommodate, beds, property type, whether the calendar is updated or not?
CONCLUSION:
Finally, we have concluded that based on our analysis:
- Between March till May, we saw a lot of listings available. It means it could be the busiest time of the year to visit Seattle
- From June till August we saw the listings prices increased, which means that fewer listings might be available during this time
- Areas such as University District, or Lake City offer cheap deals whereas the areas like Northgate, Magnolia, Cascade, etc offer some of the expensive deals.
- Most interestingly we find out that most of the people look after some of the features like bathrooms, how many people they can accommodate, beds, property type, etc while looking for a place.
If you are interested in reading more about the code they are present at my Github account link.
Also, I have written another article where I discussed all the steps in-depth of how I approached the problem. Here is the link
Please read this article and let me know your suggestions. You can reach me out on my LinkedIn account link.