Recommendation Modeling (Categorical Data)

Introduction: Statement of the problem.
Dressify is a garment company, which deals the latest fashion trends around the world. They have one of the best creative designers, and now they plan to use data science too. They have captured few data points for style, price, size, fabric type, pattern and different dress attributes over the last couple of days.
For Dressify we would like to identify the most recommended dresses based on the datapoints that they have gathered over the past days. Some of these variables will be very useful in identifying the most recommended dress, which then can be a possible sale lead.
Finally after fitting the model we will predict the Recommendation for some of their Articles, which then Dressify will recommend to its customers
Material and Methods: Description of your data and the methods you have used for the analysis.
Since, our response variable and most of the independent variables are categorical, we will use Chi-Square test of independence to retain/testing the importance of the variables in the model, and t-test for testing the importance/variation of any continuous variable with the response variable.

Description of the Data:
Style: The Style of Dress for a particular piece of clothing.
Price: Price of the Dress
Size: Mostly, dress of size medium and large has highest sell compare to small, XL and XXL.
Rating: Rating of the article.
Season: Recommendation of sale is also based on season. For example: Sale increases of woollen dress in winter compare to other season.
NeckLine: Neckline is a type of Neck for a particular piece of article.
SleeveLength: Type of Sleevelength
Waiseline: Type of Waiseline
Material: Type of Material used to create the article
FabricType: Type of Fabric used to create the article
Decoration: Material Decorated with
Pattern: Pattern on the Article
Area: Area from where the customer bought the Product

1. (2) If the p-value of a test yields p = 0.3879, one would: a....