The goal of this project is to develop a data driven model that predicts the price of an Airbnb listing based on multiple features such as location, room type, number of reviews, availability etc. Data from "Inside Airbnb", a publicly accessible website that gives information about the world wide usage of Airbnb , will be used for the goal aforementioned. The model so developed can help homeowners maximize revenue from their listings and also help prospective tenants make smart choices regarding temporary lodging.
Strengths:
The use of some pertinent questions under "Problem Description and Dataset" is a very good strategy to stimulate a reader's interest in your project.
The proposal clearly outlines the potential benefits of the project for homeowners and prospective tenants, thereby creating strong arguments in favor of developing the price prediction model.
In addition to price prediction model, the identification of features that affect the pricing also forms a lucrative part of the project as it can effectively pin point the areas that the existing homeowners can tap into to increase revenue from their properties.
Scope for Improvement:
The methodology to be used for developing the model is a bit vague and it is not immediately clear whether the end result is going to be one general price prediction model or a bunch of city specific models. I would expect the models to be city specific as they would better capture the local dynamics. Additionally, it would be interesting to note whether the set of features that influence a listing price vary from city to city.
A brief overview of the discrepancies associated with the data set and the methods for handling such discrepancies can also be included.
Adding some information regarding the learning algorithms to be used would also be helpful. It is evident that the data set is quite heterogeneous and a manager would be interested in knowing the machine learning techniques that the group is tentatively going to use to tackle this.
The goal of this project is to develop a data driven model that predicts the price of an Airbnb listing based on multiple features such as location, room type, number of reviews, availability etc. Data from "Inside Airbnb", a publicly accessible website that gives information about the world wide usage of Airbnb , will be used for the goal aforementioned. The model so developed can help homeowners maximize revenue from their listings and also help prospective tenants make smart choices regarding temporary lodging.
Strengths:
The use of some pertinent questions under "Problem Description and Dataset" is a very good strategy to stimulate a reader's interest in your project.
The proposal clearly outlines the potential benefits of the project for homeowners and prospective tenants, thereby creating strong arguments in favor of developing the price prediction model.
In addition to price prediction model, the identification of features that affect the pricing also forms a lucrative part of the project as it can effectively pin point the areas that the existing homeowners can tap into to increase revenue from their properties.
Scope for Improvement:
The methodology to be used for developing the model is a bit vague and it is not immediately clear whether the end result is going to be one general price prediction model or a bunch of city specific models. I would expect the models to be city specific as they would better capture the local dynamics. Additionally, it would be interesting to note whether the set of features that influence a listing price vary from city to city.
A brief overview of the discrepancies associated with the data set and the methods for handling such discrepancies can also be included.
Adding some information regarding the learning algorithms to be used would also be helpful. It is evident that the data set is quite heterogeneous and a manager would be interested in knowing the machine learning techniques that the group is tentatively going to use to tackle this.