About the class

Data Analysis Practical Class for D.R.M. Part 2

If you're taking this class, you're probably the one who took the first class. In class 1, I had time to learn basic statistical theory and create my own statistical tools in Python. It's safe to say that even if you've only heard this far, you have the ability to solve most data analysis problems without much difficulty.

In this session, we will go one step further and learn about methodologies for analyzing multivariate data.

Multivariate data means that an object has multiple characteristics. For example, a person named me has various characteristics such as age, age, gender, and occupation. This is multivariate data.

When to use multivariate data?

Imagine doing marketing with maximum impact when the marketing budget is limited.

Can we solve this with historical data? Of course it's possible.

First, collect all the data from past marketing activities. (Using Pandas learned in class 1)

List all the various characteristics (age, gender, occupation, number of purchases, etc.) of customers in one unit. After that, customers who have responded to marketing are divided according to their level of response, made into a single variable, and clustering is carried out. (Using hierarchical clustering)

Once the cluster is over, extract the characteristics of customers who frequently respond to marketing. (Using Decision Trees)

Marketing is carried out by selecting only customers that match these characteristics.

Aggregate marketing performance.

Check if it's really cost effective with the past.

So the multivariate data analysis is over. What do you think?

All of the things described above have been included in the lesson.

I'm confident that if you listen to this course, you'll be able to solve problems similar to those above easily!

Course effects

• You will be able to analyze multivariate data.

• You will be able to verify effectiveness (marketing, recommendations, etc.).

• No matter what you do, if you have data, you'll be able to make better decisions based on this.

Recommended target

• Those who want to take De.R.M. Class 1 and get better at data analysis

• Those who want to use machine learning for data analysis

Notes before taking the course

• Same as class 1. If you only have a laptop and internet, you can take courses and practice.

N reasons why this class is special

I'm not going to make it difficult to explain. It takes a practical approach so that it can be used intuitively and immediately.

❶ What is the difference between principal component analysis and factor analysis?

• Is it possible to divide the typical characteristics that customers have into some categories? Principal component analysis

• What characteristics do customers have in common? Factor analysis

❷ Let's group customers by characteristics! clustering

• How many typical groups are our customers divided into?

• Is it possible to share price-sensitive customers with non-military customers?

• What should I do for targeted marketing by selecting customers who will respond to marketing?

• Clustering is the answer!

❸ What are the characteristics of customers in this group? Decision tree!

• Machine learning is good at learning patterns, but it can also be used to interpret patterns.

• If customers are grouped through clustering, the characteristics of that group can be extracted into a decision tree.