Impressions etc. because I attended "Data Science Exercise for Working People (Statistics Bureau)"
The 3rd Ministry of Internal Affairs and Communications Data Science Online Course "Data Science Exercises for Working People" is now open.
In this course, you will be able to learn practical data analysis methods in business through video lectures and exercises using excel.
This time, I have finished taking all the lectures, so I will summarize the lecture outline and impressions.Since I had basic knowledge (although it is really the basics of the basics), I managed to answer all the exercises correctly.
Target course
"Data Science Exercises for Working People" provided by Statistics Bureau, Ministry of Internal Affairs and Communications
http://gacco.org/stat-japan2
Supplementary materials for this course (4 pages in A135 size) are also on sale.You can buy it officially from amazon or the Japan Statistical Association.
Data Science Exercises for Working People Official Study Note Revised Edition
If you missed the class, why not wait until the next class or see here.
Overview
Total study time: (XNUMX time XNUMX-XNUMX minutes x XNUMX-XNUMX lectures) x XNUMX weeks
Week 1: What is Data Science?
What is "data science"?
Required background, required skills / knowledge
How to proceed and method of analysis
Week 2: Analytical concepts and examples
Understand and compare data on practical business issues.
Week 3: Specific method of analysis
Cross tabulation
Scatter plot and correlation
How to read and interpret time series data
Week 4: Report Business Forecasts and Analysis Results
Regression analysis and model evaluation
Report of analysis results
Typical methods such as prediction and classification and usage situations
Week 5: To Realize Data Science in Business
Review, summary
Problem-solving case study based on data analysis
Various data scientists
Key points for realizing data science in a company
Impressions
Course level
Impression that many of the contents were basic.It seems that it is supposed to be taken after "Introduction to Data Science for Working People".
However, if you have knowledge of basic statistics such as median, correlation coefficient, variance, and regression analysis, there was no problem even if you did not receive "introduction".
About the contents
As it is said to be an exercise, it is premised that you have basic knowledge, and the content is to set an example of how to proceed when actually analyzing data in a business setting.
The teachers in charge of each lecture were such members as experts involved in data science, university professors, and staff of the Statistics Bureau of the Ministry of Internal Affairs and Communications.It was very refreshing and enjoyable for me as a learner to hear the parables that the teachers seemed to have actually faced in the field.
About exercises
It was basically an exercise rather than a lecture. I will solve it with excel, but I thought that it would be okay if I solved it with R or python for those who are starting data science in earnest.
On the other hand, I was reminded that I can do advanced things with excel.I didn't know that time series analysis could break down trends, seasonal fluctuations, and irregular fluctuations ...
There is a test every week, but I got a perfect score because the lecture was polite.I am looking forward to issuing a certificate of completion.
Try to solve it with Python
When I want to start data science / machine learning in earnest, I still want to flexibly analyze data with python and R.So I tried to solve it with python as well as learning programming.
Click here for the code.If you would like to practice python in the same way, please use it.
https://github.com/echomint/Data-science-practice-for-workers
It took me a long time to solve it while learning the basic code and data.
For routine analysis, GUI software such as Excel is sufficient.
If you try to do more complicated analysis, fine-tuning the prediction model, grasping the inside of the model, programming will be more flexible and you can handle it, and I think that it was good in terms of experiencing the basics of analysis with python. ..