IDS_1주차: IDS란?

데이터 마이닝, data mining 

: discovering patterns in large data sets by using machine learning, statistics, and database systems

Data Science,DS 

: fourth paradigm of science

 - Classification / Prediction

 - Correlation

 - Clustering

 - Rule finding

Purposes of using DS

- Productivity

- Economic cost saving

- Quality improvement

- Safety & Environment

 

Machine Learning: find a function that can predict the output from a new input

                              Y = f(X) such that rule, pattern, knowledge

Machine Learning 

(divided by labels existed) 

- Supervised Learning, divided by type of Y (Y = f(X))

          -  Classification     

          -  Prediction

 +) Semi-Supervised Learning (y = f(X))

- Unsupervised Learning

          -  Clustering (X ----> K)

          -  Association Analysis  (X <-----> X) 

   +) Self-Supervised Learning (X ------> X' , Y = f(X'))

(divided by purpose of learning)

- Reinforcement Learning

  : Times series (Yt-k = f(Yt))

 

Inductive(귀납적) VS Deductive(연역적)

- Induction : specific to general (ec. A dies, B dies , C dies ~ so that Everybody dies.)

   Cases(data) -> Theroy

- Deduction: general to specific (ec. Every man dies and Socrates is a man ~ so that Socrates dies.)

  Theroy -> Cases