There has been a high demand for online food orders after the introduction of Swiggy and Zomato in the market. Food delivery companies use your buying habits to make the delivery process faster. The food order prediction system is one of the useful techniques these companies can use to make the entire delivery process fast. In this article, I will take you through the task of Online Food Order Prediction with Machine Learning using Python.
Online Food Order Prediction: Use Case
After implementing the complete online food delivery system, companies like Swiggy and Zomato will always try to improve them. The main objective of these companies toward their customers is to deliver the food at the right time. To deliver the food faster, these companies identify areas where the demand for online food orders is high and employ more delivery partners in those locations. It helps deliver food faster in areas with more orders.
These companies have so much data about their customers that they now know the food ordering habits of all customers. With this data, they can also predict whether or not a customer will order again from their app. It is a good technique for identifying areas, families and customer types with more business opportunities.
I hope you now have understood the use case of online food order prediction systems. In the section below, I will take you through how you can train a Machine Learning model to predict online food orders from a particular customer.
Online Food Order Prediction using Python
Now let’s start with the task of online food order prediction with machine learning. You can download the dataset I am using for this task from here. Let’s start with importing the necessary Python libraries and the dataset:
import numpy as np
import pandas as pd
import numpy as np
import plotly.express as px
import plotly.graph_objects as go
import matplotlib.pyplot as plt
import seaborn as sns
sns.set_theme(style="whitegrid")
data = pd.read_csv("onlinefoods.csv")
print(data.head())Age Gender Marital Status Occupation Monthly Income \ 0 20 Female Single Student No Income 1 24 Female Single Student Below Rs.10000 2 22 Male Single Student Below Rs.10000 3 22 Female Single Student No Income 4 22 Male Single Student Below Rs.10000 Educational Qualifications Family size latitude longitude Pin code \ 0 Post Graduate 4 12.9766 77.5993 560001 1 Graduate 3 12.9770 77.5773 560009 2 Post Graduate 3 12.9551 77.6593 560017 3 Graduate 6 12.9473 77.5616 560019 4 Post Graduate 4 12.9850 77.5533 560010 Output Feedback Unnamed: 12 0 Yes Positive Yes 1 Yes Positive Yes 2 Yes Negative Yes 3 Yes Positive Yes 4 Yes Positive Yes
So the dataset contains information like:
- the age of the customer
- marital status of the customer
- occupation of the customer
- monthly income of the customer
- educational qualification of the customer
- family size of the customer
- latitude and longitude of the location of the customer
- pin code of the residence of the customer
- did the customer order again (Output)
- Feedback of the last order (Positive or Negative)
Let’s have a look at the information about all the columns in the dataset:
print(data.info())
<class 'pandas.core.frame.DataFrame'> RangeIndex: 388 entries, 0 to 387 Data columns (total 13 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 Age 388 non-null int64 1 Gender 388 non-null object 2 Marital Status 388 non-null object 3 Occupation 388 non-null object 4 Monthly Income 388 non-null object 5 Educational Qualifications 388 non-null object 6 Family size 388 non-null int64 7 latitude 388 non-null float64 8 longitude 388 non-null float64 9 Pin code 388 non-null int64 10 Output 388 non-null object 11 Feedback 388 non-null object 12 Unnamed: 12 388 non-null object dtypes: float64(2), int64(3), object(8) memory usage: 39.5+ KB None
Now let’s move to the analysis of this data. I will start by looking at the online food order decisions based on the age of the customer:
plt.figure(figsize=(15, 10))
plt.title("Online Food Order Decisions Based on the Age of the Customer")
sns.histplot(x="Age", hue="Output", data=data)
plt.show()
We can see that the age group of 22-25 ordered the food often again. It also means this age group is the target of online food delivery companies. Now let’s have a look at the online food order decisions based on the size of the family of the customer:
plt.figure(figsize=(15, 10))
plt.title("Online Food Order Decisions Based on the Size of the Family")
sns.histplot(x="Family size", hue="Output", data=data)
plt.show()
Families with 2 and 3 members are ordering food often. These can be roommates, couples, or a family of three.
Let’s create a dataset of all the customers who ordered the food again:
buying_again_data = data.query("Output == 'Yes'")
print(buying_again_data.head())Age Gender Marital Status Occupation Monthly Income \ 0 20 Female Single Student No Income 1 24 Female Single Student Below Rs.10000 2 22 Male Single Student Below Rs.10000 3 22 Female Single Student No Income 4 22 Male Single Student Below Rs.10000 Educational Qualifications Family size latitude longitude Pin code \ 0 Post Graduate 4 12.9766 77.5993 560001 1 Graduate 3 12.9770 77.5773 560009 2 Post Graduate 3 12.9551 77.6593 560017 3 Graduate 6 12.9473 77.5616 560019 4 Post Graduate 4 12.9850 77.5533 560010 Output Feedback Unnamed: 12 0 Yes Positive Yes 1 Yes Positive Yes 2 Yes Negative Yes 3 Yes Positive Yes 4 Yes Positive Yes
Now let’s have a look at the gender column. Let’s find who orders food more online:
gender = buying_again_data["Gender"].value_counts()
label = gender.index
counts = gender.values
colors = ['gold','lightgreen']
fig = go.Figure(data=[go.Pie(labels=label, values=counts)])
fig.update_layout(title_text='Who Orders Food Online More: Male Vs. Female')
fig.update_traces(hoverinfo='label+percent', textinfo='value', textfont_size=30,
marker=dict(colors=colors, line=dict(color='black', width=3)))
fig.show()
According to the dataset, male customers are ordering more compared the females. Now let’s have a look at the marital status of the customers who ordered again:
marital = buying_again_data["Marital Status"].value_counts()
label = marital.index
counts = marital.values
colors = ['gold','lightgreen']
fig = go.Figure(data=[go.Pie(labels=label, values=counts)])
fig.update_layout(title_text='Who Orders Food Online More: Married Vs. Singles')
fig.update_traces(hoverinfo='label+percent', textinfo='value', textfont_size=30,
marker=dict(colors=colors, line=dict(color='black', width=3)))
fig.show()
According to the above figure, 76.1% of the frequent customers are singles. Now let’s have a look at what’s the income group of the customers who ordered the food again:
income = buying_again_data["Monthly Income"].value_counts()
label = income.index
counts = income.values
colors = ['gold','lightgreen']
fig = go.Figure(data=[go.Pie(labels=label, values=counts)])
fig.update_layout(title_text='Which Income Group Orders Food Online More')
fig.update_traces(hoverinfo='label+percent', textinfo='value', textfont_size=30,
marker=dict(colors=colors, line=dict(color='black', width=3)))
fig.show()
According to the above figure, 54% of the customers don’t fall under any income group. They can be housewives or students.
Now let’s prepare the data for the task of training a machine learning model. Here I will convert all the categorical features into numerical values:
data["Gender"] = data["Gender"].map({"Male": 1, "Female": 0})
data["Marital Status"] = data["Marital Status"].map({"Married": 2,
"Single": 1,
"Prefer not to say": 0})
data["Occupation"] = data["Occupation"].map({"Student": 1,
"Employee": 2,
"Self Employeed": 3,
"House wife": 4})
data["Educational Qualifications"] = data["Educational Qualifications"].map({"Graduate": 1,
"Post Graduate": 2,
"Ph.D": 3, "School": 4,
"Uneducated": 5})
data["Monthly Income"] = data["Monthly Income"].map({"No Income": 0,
"25001 to 50000": 5000,
"More than 50000": 7000,
"10001 to 25000": 25000,
"Below Rs.10000": 10000})
data["Feedback"] = data["Feedback"].map({"Positive": 1, "Negative ": 0})
print(data.head())Age Gender Marital Status Occupation Monthly Income \ 0 20 0 1 1 0 1 24 0 1 1 10000 2 22 1 1 1 10000 3 22 0 1 1 0 4 22 1 1 1 10000 Educational Qualifications Family size latitude longitude Pin code \ 0 2 4 12.9766 77.5993 560001 1 1 3 12.9770 77.5773 560009 2 2 3 12.9551 77.6593 560017 3 1 6 12.9473 77.5616 560019 4 2 4 12.9850 77.5533 560010 Output Feedback Unnamed: 12 0 Yes 1 Yes 1 Yes 1 Yes 2 Yes 0 Yes 3 Yes 1 Yes 4 Yes 1 Yes
Online Food Order Prediction Model
Now let’s train a machine learning model to predict whether a customer will order again or not. I will start by splitting the data into training and test sets:
#splitting data
from sklearn.model_selection import train_test_split
x = np.array(data[["Age", "Gender", "Marital Status", "Occupation",
"Monthly Income", "Educational Qualifications",
"Family size", "Pin code", "Feedback"]])
y = np.array(data[["Output"]])Now let’s train the machine learning model:
# training a machine learning model
from sklearn.ensemble import RandomForestClassifier
xtrain, xtest, ytrain, ytest = train_test_split(x, y,
test_size=0.10,
random_state=42)
model = RandomForestClassifier()
model.fit(xtrain, ytrain)
print(model.score(xtest, ytest))0.9487179487179487
Now let’s prepare a form to input the data of the customer and predict whether the customer will order the food again or not:
print("Enter Customer Details to Predict If the Customer Will Order Again")
a = int(input("Enter the Age of the Customer: "))
b = int(input("Enter the Gender of the Customer (1 = Male, 0 = Female): "))
c = int(input("Marital Status of the Customer (1 = Single, 2 = Married, 3 = Not Revealed): "))
d = int(input("Occupation of the Customer (Student = 1, Employee = 2, Self Employeed = 3, House wife = 4): "))
e = int(input("Monthly Income: "))
f = int(input("Educational Qualification (Graduate = 1, Post Graduate = 2, Ph.D = 3, School = 4, Uneducated = 5): "))
g = int(input("Family Size: "))
h = int(input("Pin Code: "))
i = int(input("Review of the Last Order (1 = Positive, 0 = Negative): "))
features = np.array([[a, b, c, d, e, f, g, h, i]])
print("Finding if the customer will order again: ", model.predict(features))Enter Customer Details to Predict If the Customer Will Order Again Enter the Age of the Customer: 22 Enter the Gender of the Customer (1 = Male, 0 = Female): 1 Marital Status of the Customer (1 = Single, 2 = Married, 3 = Not Revealed): 1 Occupation of the Customer (Student = 1, Employee = 2, Self Employeed = 3, House wife = 4): 1 Monthly Income: 0 Educational Qualification (Graduate = 1, Post Graduate = 2, Ph.D = 3, School = 4, Uneducated = 5): 1 Family Size: 3 Pin Code: 560010 Review of the Last Order (1 = Positive, 0 = Negative): 1 Finding if the customer will order again: ['Yes']
This is how you can train a machine learning model to predict online food orders.
Summary
So this is how you can predict whether a customer will order food online again or not. The food order prediction system is one of the useful techniques food delivery companies can use to make the entire delivery process fast. I hope you liked this article on Online Food Delivery Prediction with Machine Learning. Feel free to ask valuable questions in the comments section below.
