2024 Faculty Courses School of Engineering Undergraduate major in Mechanical Engineering
Statistical data analysis
- Academic unit or major
- Undergraduate major in Mechanical Engineering
- Instructor(s)
- Satoshi Miura
- Class Format
- Lecture/Exercise (Face-to-face)
- Media-enhanced courses
- -
- Day of week/Period
(Classrooms) - 5-8 Thu
- Class
- -
- Course Code
- MEC.B221
- Number of credits
- 110
- Course offered
- 2024
- Offered quarter
- 3Q
- Syllabus updated
- Mar 17, 2025
- Language
- Japanese
Syllabus
Course overview and goals
The Genearal Instrctuve Objective is to study the data processing method using statistics, optimzation and machine learning by the theoreotical and pratical method.
Course description and aims
The specific behavioral objectives of this course are as follows
To be able to understand, explain, and practice basic statistical processing for data variability.
To be able to understand, explain, and practice basic optimization calculations.
Understand, explain, and practice basic machine learning.
Keywords
Machine learning, AI, Statistics, Probability, Disperson, Data science, Deep learning, Artificial Intelligence, Optimization
Competencies
- Specialist skills
- Intercultural skills
- Communication skills
- Critical thinking skills
- Practical and/or problem-solving skills
- This course corresponds to the following learning objectives: 6. developmental expertise in mechanical engineering 7. the ability to utilize specialized knowledge to solve new problems and make creative proposals.
Class flow
After the lectures, students will connect to the Internet on their own PCs and practice programming. Please prepare for and review each lecture.
Course schedule/Objectives
| Course schedule | Objectives | |
|---|---|---|
| Class 1 | Introduction | Population and sample space, Bayes' theorem, etc. |
| Class 2 | Data Variability | Probability dispersion function, correlation coefficient |
| Class 3 | Interpretation of Variation | Probability distribution, mean and variance, central limit theorem |
| Class 4 | Representative data and various probability distributions | Binomial, Poisson, and normal distributions |
| Class 5 | statistical testing methods | Comparison of population means, significant differences, p-values, confidence intervals |
| Class 6 | Data-based estimation methods | Maximum Likelihood Estimation Method, Information Criterion and Model Selection Theory, etc. |
| Class 7 | data modeling | Generalized linear models, linear regression, logistic regression, etc. |
| Class 8 | Optimized data visualization methods | Dimensional Compression, Principal Component Analysis, MDS, other manifold learning |
| Class 9 | clustering approach | Hierarchical, group mean, etc., non-hierarchical, k-means, etc. |
| Class 10 | Robust Estimation Methods | Least squares, M-estimation, robust estimation using random numbers, etc. |
| Class 11 | Developmental modeling using machine learning (1) | From recent and advanced topics such as nonlinear models, Bayesian networks |
| Class 12 | Developmental modeling using machine learning (2) | From recent and advanced topics such as nonlinear models and neural networks |
| Class 13 | Data science-aware research design, etc. | DNN, CNN, RNN, GAN, Transfer learning |
| Class 14 | Final Report | Final Report |
| Class 15 | - |
Study advice (preparation and review)
To enhance effective learning, students are encouraged to spend approximately 100 minutes preparing for class and another 100 minutes reviewing class content afterwards (including assignments) for each class.
They should do so by referring to textbooks and other course material.
Textbook(s)
Distribute materials as needed.
Reference books, course materials, etc.
none
Evaluation methods and criteria
Evaluation will be based on submissions in the form of reports.
Related courses
- MEC.B231 ： Probability Theory and Statistics
- MEC.B232 ： Fundamentals of Numerical Analysis
- MEC.B201 ： Fundamentals of information and mathematical sciences
Prerequisites
This course is part of
B231.L "Probability and Statistics" and MEC.B232.L "Basic Numerical Methods
L "Probability and Statistics" and "Fundamental Numerical Methods".
L "Probability and Statistics" and the former MEC.
Students who entered before March 31, 2023 (～22B) cannot take this course if they have already earned credits in both "Probability and Statistics" and "Fundamental Numerical Methods".
1 credit of L (elective) and 1 credit of non-standard course if the student has already earned credits for either of the two subjects.
If the student has not earned both credits, 2 L (elective) credits
If both credits have not been earned, the credits will be converted at the rate of 2 L (elective) credits.
Other
MEC.B201 : Students should have taken Fundamentals of Informatics and Mathematics or have equivalent knowledge.