Practical Data Science 2

Course title

33015600

Middle-level Diploma Policy (mDP)

Program / Major	mDP	Goals
Data Science Course	DP-3・1	技術文書の作成、口頭発表、討論等のコミュニケーションができる。
Data Science Course	DP-4d・2	キャリアを見据えた高度な専門知識多様なデータを収集・分析・予測する技術を駆使し、社会に存在する実際の課題に対してエビデンスを基に解決法を考え提案できる。

Purpose of class

The purpose of this course is for students to practice analysis and proposal making for real social issues by using data science, data engineering, and AI through Project Based Learning (PBL), and to acquire the ability to carry out practical-level data analysis projects while communicating with problem proposers.

Course description

Project Based Learning (PBL) is effective for acquiring practical skills in data science, data engineering, and AI. In this exercise course, students work on real social issues and conduct analysis and proposal making based on requests from problem proposers engaged in practice by using data science, data engineering, and AI. Through group work conducted jointly with students from other years, students experience the full process of understanding the problem, preprocessing data, conducting exploratory analysis, selecting methods, developing and validating models, interpreting and reporting analytical results, preparing proposals, and giving presentations. Students in this course acquire the ability to take charge of analytical work as group members under the guidance of senior students.

Goals and objectives

Understand PBL methods and acquire data utilization skills for solving real-world problems.
Understand the importance of problem setting through communication with problem proposers.
Appropriately collect, preprocess, and conduct exploratory analysis of data.
Select analytical methods suited to the problem, and build and evaluate machine learning models.
Visualize analytical results and create evidence-based proposals.
Develop the ability to carry out practical-level data analysis projects while cooperating with fourth-year students.

Relationship between 'Goals and Objectives' and 'Course Outcomes'

	Progress explanation materials for each class	Comprehensive assignment	Total.
1.	5%	5%	10%
2.	10%	10%	20%
3.	10%	10%	20%
4.	10%	10%	20%
5.	10%	10%	20%
6.	5%	5%	10%
Total.	50%	50%	-

Evaluation method and criteria

Students who earn 60 points or higher in the overall evaluation will pass. Students will be evaluated on a 100-point scale based on progress explanation materials for each class (50%) and the comprehensive assignment (50%). The progress explanation materials will assess PBL progress, sharing of issues with problem proposers, the processes of data collection, preprocessing, and exploratory analysis, explanation of analytical results, and fulfillment of roles within the group. The comprehensive assignment will assess understanding of the real issue, selection of appropriate analytical methods, construction and evaluation of machine learning models, visualization of analytical results, and the completeness of evidence-based proposals and reports. A score of 60 indicates the level at which the minimum necessary knowledge and skills for this course have been acquired.

Language

Japanese

Class schedule

	Class schedule	HW assignments (Including preparation and review of the class.)	Amount of Time Required
1.	PBL procedures and team formation	Review the basic procedures of PBL and the roles involved in team activities in advance.	180minutes
1.	PBL procedures and team formation	Organize the role assignments within the team and the way the project will proceed.	200minutes
2.	Sharing issues with problem proposers and developing a project plan	Research the background of the problem proposer, the problem area, and possible data in advance.	180minutes
2.		Organize the contents of issue sharing, and summarize the project plan and analysis policy.	200minutes
3.	Practice in data collection and preprocessing	Review methods of data collection and the basic procedures for preprocessing.	180minutes
3.	Practice in data collection and preprocessing	Check the quality of collected data and summarize the results and issues of preprocessing.	200minutes
4.	Conducting exploratory data analysis (EDA) (1): basic aggregation	Review basic aggregation methods for exploratory data analysis.	180minutes
4.		Conduct basic aggregation and organize the characteristics of the data and key findings.	200minutes
5.	Conducting exploratory data analysis (EDA) (2): basic analysis	Review visualization and basic analysis methods in exploratory data analysis.	180minutes
5.		Conduct basic analysis and summarize features and trends related to problem solving.	200minutes
6.	Selecting analytical methods according to the problem	Research candidate analytical methods and selection criteria according to the problem.	180minutes
6.	Selecting analytical methods according to the problem	Prepare explanatory materials on the reasons for selecting the analytical method and its applicability.	200minutes
7.	Implementation of machine learning models	Review the basic procedures for implementing machine learning models.	180minutes
7.	Implementation of machine learning models	Implement a machine learning model and organize the execution results and issues.	200minutes
8.	Model evaluation and improvement (1): conducting evaluation	Review model evaluation indicators and evaluation methods.	180minutes
8.	Model evaluation and improvement (1): conducting evaluation	Conduct model evaluation and summarize the evaluation results and points for improvement.	200minutes
9.	Model evaluation and improvement (2): improvement based on evaluation	Research methods for improving models based on evaluation results.	180minutes
9.		Improve the model, compare the results before and after improvement, and discuss the findings.	200minutes
10.	Visualization of analytical results and report writing	Review methods for visualizing analytical results and structuring reports.	180minutes
10.	Visualization of analytical results and report writing	Visualize analytical results and prepare an outline of the report.	200minutes
11.	Creating data-based problem-solving proposals (1): examining proposal structure	Review the structure of data-based proposals and methods for persuasive explanation.	180minutes
11.		Examine the structure of the problem-solving proposal and organize the analytical results that support it.	200minutes
12.	Creating data-based problem-solving proposals (2): finalizing the proposal	Check the validity of the proposal and identify any missing analyses.	180minutes
12.		Finalize the proposal materials and improve the explanation for the presentation.	200minutes
13.	Final presentation: presentation of proposals	Review the presentation materials and prepare to explain the proposal and analytical evidence.	160minutes
13.	Final presentation: presentation of proposals	Organize questions and comments received during the presentation and reflect them in the comprehensive assignment.	210minutes
14.	Reflection and improvement measures for the next academic year	Prepare to reflect on the overall outcomes and issues of the project.	160minutes
14.		Based on the reflection, organize improvement measures for the next academic year.	210minutes
Total.	-	-	5300minutes