Personal loans are a lucrative revenue stream for banks. The typical interest rate of a two year loan in the United Kingdom is around 10%. This might not sound like a lot, but in September 2022 alone UK consumers borrowed around £1.5 billion!
You have been asked to work with a bank to clean and store the data they collected as part of a recent marketing campaign, which aimed to get customers to take out a personal loan. They plan to conduct more marketing campaigns going forward so would like you to set up a PostgreSQL database to store this campaign's data, designing the schema in a way that would allow data from future campaigns to be easily imported.
They have supplied you with a csv file called "bank_marketing.csv", which you will need to clean, reformat, and split, in order to save separate files based on the tables you will create. Lastly, you will write the SQL code that the bank can execute to create the tables and populate with the data from the csv files. As the bank are quite strict about their security, you'll provide the database design script as a .sql file that they can then run.
You have been asked to design a database that will have three tables:
client
client| column | data type | description |
|---|---|---|
id | serial | Client ID - primary key |
age | integer | Client's age in years |
job | text | Client's type of job |
marital | text | Client's marital status |
education | text | Client's level of education |
credit_default | text | Whether the client's credit is in default |
housing | text | Whether the client has an existing housing loan (mortgage) |
loan | text | Whether the client has an existing personal loan |
campaign
campaign| column | data type | description |
|---|---|---|
campaign_id | integer | Campaign ID |
client_id | serial | Client ID - references id in the client table |
number_contacts | integer | Number of contact attempts to the client in the current campaign |
contact_duration | integer | Last contact duration in seconds |
pdays | integer | Number of days since contact in previous campaign (999 = not previously contacted) |
previous_campaign_contacts | integer | Number of contact attempts to the client in the previous campaign |
previous_outcome | text | Outcome of the previous campaign |
campaign_outcome | integer | Outcome of the current campaign |
last_contact_date | date | Last date the client was contacted |
economics
economics| column | data type | description |
|---|---|---|
client_id | serial | Client ID - references id in the client table |
emp_var_rate | float | Employment variation rate (quarterly indicator) |
cons_price_idx | float | Consumer price index (monthly indicator) |
euribor_three_months | float | Euro Interbank Offered Rate (euribor) three month rate (daily indicator) |
number_employed | float | Number of employees (quarterly indicator) |
# Start coding...
import pandas as pd
import numpy as np# Store and print database_design
# Load the data
df = pd.read_csv("bank_marketing.csv")
# Split the data
client = df[["client_id", "age", "job", "marital",
"education", "credit_default", "housing", "loan"]]
campaign = df[["client_id", "campaign", "month", "day_of_week",
"duration", "pdays", "previous", "poutcome", "y"]]
economics = df[["client_id", "emp_var_rate", "cons_price_idx",
"euribor3m", "nr_employed"]]
# Rename, clean, create, and delete columns
## Renaming columns
client.rename(columns={"client_id": "id"},
inplace=True)
campaign.rename(columns={"duration": "contact_duration",
"previous": "previous_campaign_contacts",
"y": "campaign_outcome",
"campaign": "number_contacts",
"poutcome": "previous_outcome"},
inplace=True)
economics.rename(columns={"euribor3m": "euribor_three_months",
"nr_employed": "number_employed"},
inplace=True)
## Cleaning columns
client["education"] = client["education"].str.replace(".", "_")
client["education"] = client["education"].replace("unknown", np.NaN)
client["job"] = client["job"].str.replace(".", "")
campaign["campaign_outcome"] = campaign["campaign_outcome"].replace("yes", 1).replace("no", 0)
campaign["previous_outcome"] = campaign["previous_outcome"].replace("non_existent", np.NaN).replace("failure", 0).replace("success", 1)
## Creating new columns
campaign["month"] = campaign["month"].str.capitalize()
campaign["day_of_week"] = campaign["day_of_week"].str.capitalize()
campaign["year"] = "2022"
campaign["last_contact_date"] = campaign["year"] + "-" + campaign["month"] + "-" + campaign["day_of_week"]
campaign["last_contact_date"] = pd.to_datetime(campaign["last_contact_date"],
format="%Y-%b-%a")
campaign["campaign_id"] = 1
## Deleting columns
campaign.drop(columns=["month",
"year",
"day_of_week"],
inplace=True)
# Saving the data
## Store the DataFrames
client.to_csv("client.csv", index=False)
campaign.to_csv("campaign.csv", index=False)
economics.to_csv("economics.csv", index=False)
## Structuring the dictionary
database_design = {"client": client,
"campaign": campaign,
"economics": economics}
print(database_design)