Top Page | About This Site | JAPANESE


These Q&A are I got by E-mails etc.

I introduce without secret.

Method of Data Science

Can you estimate yield by the sampling test data?

What is the best model of for the ralationship between occurrence data and continuous data?

Since the p-value is greater than 0.05, "no difference", right?

What is the difference between Taguhi method and desighn of experiment?

What is the best number of the data?

It is not normal distribution. How should I do?

Can you make summary of the text automatically?

Do not you use moving windows for sensor data?

Do you serch strong correlation for cause and effect analysis?

If Y is quantitative, use T method. If Y is qualitative, use MT method?

Cause is this variable?

How should I do to make smalller the unevennes?

What is the best model for this data?

Do you ever make high-dimensional (multivariate) models?

Is the hypothesis wrong because the result may come earlier?

Why don't you just swap the rows and columns?

Time is not evenly spaced. How analyze?

Is it analyzed by Fourier transform?

Tool of Data Sceience

What is the tool (software) you often use?

What is the method (model) you often use?

What spec of computer I must prepare for the data analysis?

How to study

How do you read references?

This page is made from May 21 2020. I add new page if I remember old Q&A.