The project, looking forward to solving a problem for Invistico Airlines, was divided into several sections. The first step in any data science project is understanding the problem. In our case, we discovered the problem after analyzing the dataset, calculating the mean, standard deviation, min, max, etc. of features, defining the target, indexing, and retrieving data.
The most important point for understanding the business problem is using data visualization (the next step). It allows us to identify the relation between the features and the target variable by using a correlation matrix, and for the median, upper and lower quartiles, minimum and maximum values, and any outliers in the dataset, we used a box plot.
The third step is creating a profile report and visual data with Sweetviz. The report includes a summary of the data, missing values, and a summary of variables.
Then we went to data preparation by searching to see if we had missing data, outliers, or duplicate data, and then we tried to solve each problem.
The general problem we have in our data is that 45% of the customers are dissatisfied about flying with Invistico Airlines, so it's our task to find why they were dissatisfied. How can we solve this problem and help the company?
The modeling step was choosing a decision tree and an artificial neural network;
I'm excited to receive your opinions and advice.