So basically this is a kind of clustering project in which we are finding out or we are analyzing the personality of customers, like how these customers are, what are their habits and what are their views on their product, and then we'll try to segment them. So we'll try to do the same using two different approaches.So again, we'll be performing this operation with two different approaches. So now, but what is the motivation or what is the main reason we are doing this project or what is the importance? So asset and customer personality analysis is a detailed analysis of a company's ideal customers. Obviously it will help the company to grow and understand its customer and makes it easier. Basically understand it in the kind of like excel.
It understands the kind of feedback system like you are giving out or taking out the feedback from the customers. Okay, so this is a kind of feedback system which will help us to modify our product according to a specific need.
So customer customer personality analysis helps our business to modify its product based on its target customer.
So as I said, suppose we have an example like instead of spending money to market a new product to every customer in the company's database, a company can analyze which customer segment is most likely to buy the product. This is how you can proceed.
Like instead of selling this product to every other person, we can just segment a group of people whom we are very sure that these people buy our product.
So this will take this will save our time, money and improve our productivity. So now we will be performing exploratory data analysis for in this project, like we'll be performing or plotting some figures, answering questions like what is the distribution of customers, What is the education level of customers, the income of customers? We'll find out the relation between different parameters.Okay, what can be the relation between them then?
I've taken this dataset from Kaggle, so I have planned to do this project into different ways. Okay, We'll try to first perform the clustering using LBO(Elbow) method in K means clustering. There's a method called elbow method in which we find out the optimal number of cases.
Then we'll try to use or we'll see that if we can use super rich supervised learning algorithms to do this or not, basically clustering unsupervised algorithm, you don't know the results. So we'll also see if we can use the supervised algorithms in this to bring results. So finally, the timeline of the project will be something like this.
First, we'll be importing the libraries, then a little data analysis, some future engineering techniques.