With the development of IoT (Internet of Things) in recent years, a huge amount of data collected in physical space is stored in cyberspace every day as big data. In the Society 5.0 society that Japan is aiming for, we believe that utilizing this big data will bring various innovations to companies and society. In order to generate this innovation, it is necessary to mechanically process, analyze and interpret the stored big data. Therefore, what is needed is a study called “data science,” and a “data scientist” who can make full use of AI analysis technology in data science. Data scientists are one of the professions that are currently attracting attention, and many human resources are required. Why don’t you take this opportunity to study data science and become an innovator in the future society?
２．Our efforts
Toyohashi University of Technology aims to create a comprehensive educational environment that promotes the establishment of data science in manufacturing technology. Selected as a cooperating school for the “Nationwide Expansion of Mathematical and Data Science Education in Japan” project. In this project, the whole country is divided into 6 blocks, led by the Consortium for Strengthening Mathematical and Data Science Education, and 6 universities are the base schools. On the other hand, we are building an environment where we can receive mathematical and data science education. Based on the founding spirit of “science of technology,” our university aims to develop human resources who can develop data science from science to manufacturing technology.
With Kikagaku Co., Ltd., we have developed electronic teaching materials for data science that can be applied in various engineering fields, aiming to enable not only information students but also students and researchers of all systems to utilize big data in their own research fields. This teaching material consists of the basic edition (TK Basic series) and the advanced edition (TK Advance series) of data science, and combines self-study using electronic textbooks (E-Learning teaching materials) and practical exercises using data processing tools. We will expand these teaching materials not only to our university but also to other universities and companies, and based on the evaluations obtained, we will further improve and expand the teaching materials. If you would like to utilize our data science teaching materials in private companies and educational institutions, please contact us at the following e-mail add
Center for IT-based Education Secretariat: cite-office@cite.tut.ac.jp
■ TK Basic series
–
Learning theme
Contents
001
introduction
・What is artificial intelligence (AI)? ・Development flow of machine learning ・Build a Python development environment on a local PC ・Install the Python package used for machine learning
・Determine the model, determine the objective function, and find the optimum parameters
007
Python Basics 1 (Data Structures and Control Structures)
・Basics of Google Colaboratory ・Basic usage of Colab ・Features of Python ・Python basics (data structure) ・Control syntax
008
Python basics 2 (functions)
・What is a function? ・How to define a function ・Define various types of functions
009
Python basics 3 (class)
・ What is a class? ・ How to define a class ・ Class with variables ・ Class with function ・ Program management
010
Numerical calculation
・ Basics of NumPy ・ Calculation method using multidimensional array ・ Multiple regression analysis using NumPy
011
Data processing and visualization
・ Basics of data processing by Pandas ・ Basics of data visualization with Matplotlib
012
Implementation of machine learning 1 (Supervised learning: regression)
・ Implementation of multiple regression analysis ・ A method to suppress overfitting of linear regression ・ Correlation and multicollinearity problem
013
Implementation of machine learning 2 (Supervised learning: classification)
・ Understand the overall picture of classification by implementing a decision tree ・ Typical classification algorithm ・ Classification evaluation method ・ Check the evaluation index with scikit-learn
014
Implementation of machine learning 3 (hyperparameter adjustment)
・ Overview of hyperparameters and cross-validation ・ How to adjust hyperparameters
・Basics of neural networks ・Calculation of forward propagation
002
Neural Network Mathematics 2 (Backpropagation)
・Gradient descent method ・Mini-batch learning ・Calculation of parameter update amount
003
Implementation of neural network 1 (Classification)
・Basics of TensorFlow ・Learning classification model by TensorFlow ・Saving and inferring trained models
004
Implementation of neural network 2 (regression)
・Data set preparation ・Model definition ・Selection of objective function and optimization method ・Model learning ・Evaluation of prediction accuracy
005
Image processing and deep learning
・Basics of image processing ・Overview and structure of Convolutional Neural Network (CNN)
006
Implementation of image classification
・Basics of image processing ・Implementation of image classification ・Flow of forward propagation of CNN model
007
Series modeling and deep learning
・Recurrent neural network (RNN) ・Long-Short Term Memory (LSTM) ・Attention
008
Time series analysis
・Handling of time series data ・Implementation of binary classification of stock price rise / fall
009
Natural language processing and deep learning
・ Morphological analysis with MeCab ・ Feature conversion with Bag of Words ・ Implementation of document classification
010
Implementation of machine translation
・ Overview of Seq2Seq ・ Sentence generation with Seq2Seq
４．Data science courses using our original teaching materials
Data science courses using our original data science teaching materials include “Data Science Exercise Basics”, “Data Science Exercise Applications”, and “Mathematical / Data Science Exercise Basics”. The data science practice basics and the mathematical / data science practice basics have the same content, and the “TK Basic series” is used as a teaching material, and the “data science practice application” uses the “TK Advance series”.
“Mathematics / Data Science / AI Education Program Accreditation System” has started in order to encourage the efforts of mathematics, data science, and AI education at each university and college of technology in collaboration with the Cabinet Office, Ministry of Education, Culture, Sports, Science and Technology, and Ministry of Economy, Trade and Industry.
We are planning to apply for the Mathematical / Data Science / AI Education Program of our university to the literacy level of the “Mathematical / Data Science / AI Education Program Accreditation System”, Literacy Plus. We are also preparing to apply for the application basic level certification system.
