[본 교과목은 "산업인공지능(산업AI)프로그램 기초공통분야" 과목임]
Data science requires integration of statistical knowledge, computing skill, data-related technique, and communication skills. This course is to develop the appropriate skills to handle complex data science projects and “think with data” for better decisions. Data analysts must be able to have a right question, obtain data to address it, model and explore performing reproducible results. Programming technique for R and Python is developed throughout this course.

What we learn in this course
● data extraction techniques (data wrangling, SQL query, data features)
● modern data visualization approaches (multiple graphics, spatial data)
● boosting, regularization and clustering
● refined computational approaches (automation machine learning)

Prerequisite : Statistics course (IMEN272, Math231, Math230) or equivalent
Recommend (but not required) : Basic level of programming with R (or Python), Data mining related courses (IMEN 382, IMEN 472, IMEN 473).

Attendance & short test* 15% (short test given at the end of the semester)
Homework (quiz) 30%
midterm project 25%
final project 30%
Total 100

[Reference]
● Mastering machine learning with R, 3rd Cory Lesmeister, 2019, Packt Publishing.
● Rohan Chopra et al,, Mohamed Noordeen Alaudeen, Master Data Science with Python, 2019
● Ani Adhikari, John DeNero, Computational and Inferential Thinking: The Foundations of Data Science (free web)

1. What is Data Science / Data Science, Open source, Github
2. Computing tools for R / R&RStudio basics, Intro in R, Intro about visualization
3. Data Wrangling & Data preparation /Essencial techniques for data management (dylpr in R)
4. Statistical modeling / sampling, bootstrap, outlier model with confounding factor
5. Data handling with PYthon / Python basics, Data loading, handling(pandas)
6. Database & SQL / SQL concept and practice
7. Data visualization / matplotlib in Python. Building google map
8. Data preprocessing / Data preprocessing with scikit learn
9. Supervised learning : Classifiers / Tuning parameters, Ensemble methods
10. Evaluating methods / Cross-validation, Confusion matrix, ROC curve
11. Unsupervised learning : Clustering / Parallelizing hierachical clustering
12. Data features / Feature extraction method, Linear/nonlinear/deep learning
13. Unbalancing data/Synthetic Minority Over-sampling
14. Regularization, Automation machine learning / LASSO and ElasticNet, Auto ML

* week5-week8 : lecture by Prof. Song MinSeok
* Project presentation will be scheduled in class.

1. 대면수업 : 다양한 소스의 공유데이터를 다루고 데이터사이언티스로서의 분석력을 향상시킬수 있도록 강의진행합니다.
2. 챗GPT 활용 : 데이터분석 프로그래밍에 활용(오류 해결 및 분석기법 예제 코드 등)하고, 학습과 지식탐색을 위한 도구로 사용할 수 있습니다.
3. 과제 및 퀴즈는 PLMS에 과제마감일까지 word 혹은 ppt파일 형태로 올리고, 늦은제출에는 감점(10%-20%)이 적용됩니다.
3. 강의조교 : 한유정(huj1259@postech.ac.kr)

★ More information about the course:
* R and Python code corresponding for each subjects in main text book are given.
* For your own project or homework, you may use R or Python whatever you are comfortable.
* Recommend to try R for Python users, and to try Python for R users.
* Choose your specific data, and work these real data (public data) on all steps in the course