Gain insights into performing predictive data analysis using Python. Learn about tools like NumPy, Pandas, Matplotlib, and Seaborn, and applying techniques to real-world financial and advertisement projects.

seaborn.tar.gz

seaborn

In this course, you will learn how to perform predictive data analysis using Python. The ideal audience is those who want to start their careers as data analysts. The main goal of this course is to show you how to use statistics to draw useful insights from data which can help in predicting future behavior or patterns.

Beyond that, you’ll learn all the tools of the trade that data scientists use everyday including: NumPy, Pandas, Matplotlib, and Seaborn. You’ll learn not only how to extract meaningful insights from data, but you’ll also learn how to create stunning visualizations that you can use for reports.

Various datasets of real-world scenarios are used in each lesson to get you accustomed to handling any type of data. At the end of the course, you will work on two real-world projects that demonstrate how data analysis techniques are being used in the financial and advertisement sector to generate revenue.

Predictive Data Analysis with Python

# Exploring the data

The data files for both *October* &and*November* are very large in size so for this exercise, the data for only *October* will be selected and used. Let's review and analyze what data is stored in which format.

The following code reads the *CSV* data file and prints the `DataFrame`.

> __Note__: The code in this chapter is not runnable due to memory constraints. Please run them locally along with each lesson.

import pandas as pd

df = pd.read_csv("2019-Oct.csv") # Reading the data from file

print(df)

According to the output, there are **nine** columns in the `DataFrame`, which are described below:

* `event_time`: The exact time when the activity occurred by a user

* `event_type`: The type of activity occurred; there are three types in our case, i.e, *view*, *cart*, and *purchase*

* `product_id`: The unique ID of a particular product

* `category_id`: The unique ID of the category to which the product belongs to

* `category_code`: The unique category code to which the product belongs to

* `brand`: The brand name of the selected product

* `price`: The price of the selected product

* `user_id`: The unique ID of the user

* `user_session`: The unique ID generated every time a user visits the site. It is different for every visit of a particular user


# Brand analysis

A _brand_ is a term that differentiates one product from another.
In this analysis, we will review whether people like to purchase products with a popular brand or a product without a brand.

For this analysis, only the products actually bought by the users will be considered. In our dataset, the products which have no brand are given a `NaN` value.  
This will be done in two steps:

1. Separate the original `DataFrame` into two DataFrames. One with all the products with brands and one with all the products without brands.

2. Fetch all those rows from the two `DataFrames` where the `event_type` value is `purchase`.

As a final result, two `Dataframes` will be obtained containing the brand products with and without, that was purchased.

In this lesson, the impact of a brand name on consumer purchasing behavior is discussed.

Getting Started

Numpy for Python

Pandas for Python

Statistics for Data Analysis

Data Wrangling

Visualizing the Data

Data Scraping

Project #1

Project #2

Conclusion

Predictive Data Analysis Exam

Section 1: Brand Analysis

Exploring the data