Book Image

Data Science for Marketing Analytics

By : Tommy Blanchard, Debasish Behera, Pranshu Bhatnagar
Book Image

Data Science for Marketing Analytics

By: Tommy Blanchard, Debasish Behera, Pranshu Bhatnagar

Overview of this book

Data Science for Marketing Analytics covers every stage of data analytics, from working with a raw dataset to segmenting a population and modeling different parts of the population based on the segments. The book starts by teaching you how to use Python libraries, such as pandas and Matplotlib, to read data from Python, manipulate it, and create plots, using both categorical and continuous variables. Then, you'll learn how to segment a population into groups and use different clustering techniques to evaluate customer segmentation. As you make your way through the chapters, you'll explore ways to evaluate and select the best segmentation approach, and go on to create a linear regression model on customer value data to predict lifetime value. In the concluding chapters, you'll gain an understanding of regression techniques and tools for evaluating regression models, and explore ways to predict customer choice using classification algorithms. Finally, you'll apply these techniques to create a churn model for modeling customer product choices. By the end of this book, you will be able to build your own marketing reporting and interactive dashboard solutions.

Related Content you might be interested in

Current Title:

Small book image
Data Science for Marketing Analytics
Tommy Blanchard | Debasish Behera | Pranshu Bhatnagar
Mar, 2019 | 420 pages
Small book image
The Data Science Workshop
Robert Thas John | Anthony So | Thomas V Joseph | Andrew Worsley | Samuel Asare
Jan, 2020 | 818 pages
Small book image
The Data Science Workshop
Robert Thas John | Anthony So | Thomas V Joseph | Andrew Worsley | Samuel Asare
Aug, 2020 | 824 pages
Small book image
Applied Supervised Learning with Python
Benjamin Johnston | Ishita Mathur
Apr, 2019 | 404 pages
Table of Contents (12 chapters)
Data Science for Marketing Analytics
Data Science for Marketing Analytics
Preface
Preface
Free Chapter
1
Data Preparation and Cleaning
Data Preparation and Cleaning
Introduction
Data Models and Structured Data
pandas
Data Manipulation
Summary
2
Data Exploration and Visualization
Data Exploration and Visualization
Introduction
Identifying the Right Attributes
Generating Targeted Insights
Visualizing Data
Summary
3
Unsupervised Learning: Customer Segmentation
Unsupervised Learning: Customer Segmentation
Introduction
Customer Segmentation Methods
Similarity and Data Standardization
k-means Clustering
Summary
4
Choosing the Best Segmentation Approach
Choosing the Best Segmentation Approach
Introduction
Choosing the Number of Clusters
Different Methods of Clustering
Evaluating Clustering
Summary
5
Predicting Customer Revenue Using Linear Regression
Predicting Customer Revenue Using Linear Regression
Introduction
Understanding Regression
Feature Engineering for Regression
Performing and Interpreting Linear Regression
Summary
6
Other Regression Techniques and Tools for Evaluation
Other Regression Techniques and Tools for Evaluation
Introduction
Evaluating the Accuracy of a Regression Model
Using Regularization for Feature Selection
Tree-Based Regression Models
Summary
7
Supervised Learning: Predicting Customer Churn
Supervised Learning: Predicting Customer Churn
Introduction
Classification Problems
Understanding Logistic Regression
Creating a Data Science Pipeline
Modeling the Data
Summary
8
Fine-Tuning Classification Algorithms
Fine-Tuning Classification Algorithms
Introduction
Support Vector Machines
Decision Trees
Random Forest
Preprocessing Data for Machine Learning Models
Model Evaluation
Performance Metrics
Summary
9
Modeling Customer Choice
Modeling Customer Choice
Introduction
Understanding Multiclass Classification
Class Imbalanced Data
Summary
Appendix
Appendix
Chapter 1: Data Preparation and Cleaning
Chapter 2: Data Exploration and Visualization
Chapter 3: Unsupervised Learning: Customer Segmentation
Chapter 4: Choosing the Best Segmentation Approach
Chapter 5: Predicting Customer Revenue Using Linear Regression
Chapter 6: Other Regression Techniques and Tools for Evaluation
Chapter 7: Supervised Learning: Predicting Customer Churn
Chapter 8: Fine-Tuning Classification Algorithms
Chapter 9: Modeling Customer Choice

Chapter 5: Predicting Customer Revenue Using Linear Regression

Activity 8: Examining Relationships between Storefront Locations and Features about their Area

  1. Load the data from location_rev.csv and then take a look at it:

    import pandas as pd

df = pd.read_csv('location_rev.csv')
df.head()

  2. Use seaborn's pairplot function to visualize the data and its relationships:

    import seaborn as sns
%matplotlib inline

sns.pairplot(df)

  3. Use correlations to investigate the relationship between the different variables and the location revenue:

    df.corr()

The resulting correlations should make sense. The more competitors in the area, the lower the revenue of a location, while the median income, loyalty members, and population density are all positively related. A location's age is also positively correlated with revenue, indicating that the longer a location is open, the better known it is and the more customers it attracts (or perhaps, only locations that do well last a long time).

Activity 9: Building a Regression Model to Predict Storefront Location Revenue

  1. Import the data from location_rev.csv and view the first few rows:

    import pandas as pd
df = pd.read_csv('location_rev.csv')
df.head()

  2. Create a variable, X, with the predictors in it, and store the outcome (revenue) in a separate variable, y:

    X = df[['num_competitors',
       'median_income',
       'num_loyalty_members',
       'population_density',
       'location_age'
       ]]
y = df['revenue']

  3. Split the data into training and test sets. Use random_state = 100:

    from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X, y, random_state = 100)

  4. Create a linear regression model and fit it on the training data:

    from sklearn.linear_model import LinearRegression

model = LinearRegression()
model.fit(X_train,y_train)

  5. Print out the model coefficients:

    model.coef_

  6. Print out the model intercept:

    model.intercept_

  7. Produce a prediction for a location that has 3 competitors; a median income of 30,000; 1,200 loyalty members; a population density of 2,000; and a location age of 10:

    single_location = pd.DataFrame({
    'num_competitors': [3],
    'median_income': [30000],
    'num_loyalty_members': [1200],
    'population_density': [2000],
    'location_age': [10]
})

model.predict(single_location)

  8. Plot the model predictions versus the true values on the test data:

    import matplotlib.pyplot as plt
%matplotlib inline

plt.scatter(model.predict(X_test),y_test)
plt.xlabel('Model Predictions')
plt.ylabel('True Value')
plt.plot([0, 100000], [0, 100000], 'k-', color = 'r')
plt.show()

  9. Calculate the correlation between the model predictions and the true values of the test data:

    from scipy.stats.stats import pearsonr

pearsonr(model.predict(X_test),y_test)

The first number shows an extremely high correlation value (just over 0.9, where 1.0 would be a perfect correlation). The second number shows an extremely small p-value, indicating that it's very unlikely that this correlation is due to chance. Taken together, this indicates that our model is working very well on the test data.

Previous Section
End of Section
Next Section