Conde Nast Data Science Hiring Challenge
Preferred UI theme
You are working in an e-commerce company. The management of the company has recently changed the UI of the website. Your company has conducted a survey among the users to collect their views on the new UI update.
Task
You are required to build a machine learning model that can predict the preferred UI theme, given a user’s UI engagement information.
Dataset description
The dataset folder contains the following files:
- train.csv: 15150 x 16
- test.csv: 1850 x 15
- sample_submission.csv: 5 x 2
The columns provided in the dataset are as follows:
Column name | Description |
---|---|
CustomerID | Represents a unique identification of a user |
Age | Represents the age of the user |
Gender | Represents the gender of the user |
City | Represents the city in which the user lives |
State | Represents the state in which the user lives |
No_of_orders_placed | Represents the total number of orders placed by a customer |
Sign_up_date | Represents the date when a customer started using the website. |
Last_order_placed_date | Represents the last date when the customer placed the order |
is_premium_member | Represents whether a customer is a premium member or not. ( 0 or 1) |
Women’s_Clothing | Represents user’s engagement score in Women’s_Clothing section ( 0 to 10 ) |
Men’s_Clothing | Represents user’s engagement score in Men’s_Clothing section( 0 to 10 ) |
Kid’s_Clothing | Represents user’s engagement score in Kid’s_Clothing section ( 0 to 10 ) |
Home_&_Living | Represents user’s engagement score in Home_&_Living section ( 0 to 10 ) |
Beauty | Represents user’s engagement score in Beauty products section ( 0 to 10 ) |
Electronics | Represents user’s engagement score in Electronics products section( 0 to 10 ) |
Preferred_Theme | Represents the preferred theme ( Old_UI or New_UI ) |
Evaluation metric
score = 100*(metrics.roc_auc_score(actual, predicted, average= "weighted" ))
Result submission guidelines
- The index is "CustomerID" and the target is the "Preferred_Theme" column.
- The submission file must be submitted in .csv format only.
- The size of this submission file must be 1850 x 2.
Note: Ensure that your submission file contains the following:
- Correct index values as per the test file
- Correct names of columns as provided in the sample_submission.csv file
Instructions:
- Click Download dataset to download the dataset.
- Solve the problem in your local environment.
- Save the submission in a .csv file.
- Click Upload File (under the Upload File section) to upload your prediction file (.csv).
- Click Upload File (under the Upload Source Code section) to upload your .ipynb file along with any presentation file.
- Add any instructions or comments in the Your Answer section.
- Click Submit.