You are on page 1of 1

Machine Learning Assignment

- Download winequality-red_cl.csv from the class website.


- The data are related to red wine variants of the Portuguese "Vinho Verde" wine.
- The dataset has 1599 rows and 13 columns.

1 - fixed acidity (결합산): 와인의 산도를 제어


2 - volatile acidity (휘발산): 와인의 향에 연관
3 - citric acid (구연산): 와인의 신선함을 유지시켜주는 역할
4 - residual sugar (잔여 설탕): 와인의 단맛에 연관
5 – chlorides (염소): 와인의 짠맛과 신맛을 좌우하는 성분
6 - free sulfur dioxide (독립 이산화황): 와인의 보관도와 연관
7 - total sulfur dioxide (총 이산화황): 와인의 보관도와 연관
8 – density (밀도) : 바디의 높고 낮음을 표현하는 와인의 바디감과 연관
9 - pH (산성도) : 와인의 신맛의 정도
10 – sulphates (황산염): 와인의 보관도와 연관
11 - alcohol
12 - quality (score between 0 and 10)
13 – decision (good vs not good)

Using the k-NN (K = 5) algorithm,


(1) classify wines into good and not good. (target variable: decision)
(2) predict the quality of wines (target variable: quality)

Use all variables, except free sulfur dioxide and sulphates, in the dataset as predictors. Partition the
data into training (70%) and validation (30%). For (1), obtain performance metrics (i.e., confusion
matrix) of the model in applying it to the validation data. For (2), obtain performance metrics (i.e.,
RMSE) of the model in applying it to the validation data.

Submit a 6-page Word document that includes:

For (1),
(a) Process view after all the necessary steps are completed
(b) Results view of confusion matrix
(c) Results view (only first page) of ExampleSet (Apply Model) showing the prediction of the model

For (2),
(a) Process view after all the necessary steps are completed
(b) Results view of RMSE
(c) Results view (only first page) of ExampleSet (Apply Model) showing the prediction of the model

Use “Alt + Print Screen” to copy screen shots of the Process and Result views. Paste the screen
shots into the Word document and submit the Word file to LearnUS.

You might also like