# text result

## tf-idf 4000 word
### Category: openness

[[0 4 0]
[0 5 0]
[0 6 0]]
precision recall f1-score support

0 0.00 0.00 0.00 4
1 0.33 1.00 0.50 5
2 0.00 0.00 0.00 6

accuracy 0.33 15
macro avg 0.11 0.33 0.17 15
weighted avg 0.11 0.33 0.17 15

0.3333333333333333
**average k-fold accuracy:** 0.3004945054945055

### Category: outdoor

[[1 5]
[4 5]]
precision recall f1-score support

0 0.20 0.17 0.18 6
1 0.50 0.56 0.53 9

accuracy 0.40 15
macro avg 0.35 0.36 0.35 15
weighted avg 0.38 0.40 0.39 15

0.4
**average k-fold accuracy:** 0.5061904761904763

## Chi-square 500
### Category: openness

[[0 3 1]
[0 4 1]
[1 5 0]]
precision recall f1-score support

0 0.00 0.00 0.00 4
1 0.33 0.80 0.47 5
2 0.00 0.00 0.00 6

accuracy 0.27 15
macro avg 0.11 0.27 0.16 15
weighted avg 0.11 0.27 0.16 15

0.26666666666666666
0.3478296703296703

### Category: outdoor

[[4 2]
[0 9]]
precision recall f1-score support

0 1.00 0.67 0.80 6
1 0.82 1.00 0.90 9

accuracy 0.87 15
macro avg 0.91 0.83 0.85 15
weighted avg 0.89 0.87 0.86 15

0.8666666666666667
0.7642857142857142

## chi-square 50

### Category: openness

[[0 2 2]
[0 3 2]
[1 2 3]]
precision recall f1-score support

0 0.00 0.00 0.00 4
1 0.43 0.60 0.50 5
2 0.43 0.50 0.46 6

accuracy 0.40 15
macro avg 0.29 0.37 0.32 15
weighted avg 0.31 0.40 0.35 15

0.4
0.39645604395604395

### Category: outdoor

[[4 2]
[1 8]]
precision recall f1-score support

0 0.80 0.67 0.73 6
1 0.80 0.89 0.84 9

accuracy 0.80 15
macro avg 0.80 0.78 0.78 15
weighted avg 0.80 0.80 0.80 15

0.8
0.7104761904761905

# Summary:
tf-idf 4000:
1. openness
* Sample accuracy: 0.3333333333333333
* average k-fold accuracy: 0.3004945054945055
2. outdoor
* Sample accuracy: 0.4
* average k-fold accuracy: 0.5061904761904763

Chi-square 500:
1. openness
* Sample accuracy: 0.26666666666666666
* average k-fold accuracy: 0.3478296703296703
2. outdoor
* Sample accuracy: 0.8666666666666667
* average k-fold accuracy: 0.7642857142857142

Chi-square 50:
1. openness
* Sample accuracy: 0.4
* average k-fold accuracy: 0.39645604395604395
2. outdoor
* Sample accuracy: 0.8
* average k-fold accuracy: 0.7104761904761905

count Chi-square 500:
1. openness
* Sample accuracy: 0.3
* average k-fold accuracy: 0.3166666666666666
2. outdoor
* Sample accuracy: 0.5
* average k-fold accuracy: 0.576310715609937

count Chi-square 50:
1. openness
* Sample accuracy: 0.4
* average k-fold accuracy: 0.33689655172413796
2. outdoor
* Sample accuracy: 0.43333333333333335
* average k-fold accuracy: 0.5252836484983314

# Model difference(tf-idf chi 500)

## Personality
* Openess:
* random forest = 0.3466666666666667
* svm = 0.3704597701149425
* xgboost = 0.33999999999999997

* conscientiousness:
* random forest = 0.2949430584247047
* svm = 0.3704597701149425
* xgboost = 0.2691686530006886

* extraversion:
* random forest = 0.3579569892473119
* svm = 0.3704597701149425
* xgboost = 0.369715821812596

* aggreablness:
* random forest = 0.298963133640553
* svm = 0.3704597701149425
* xgboost = 0.30595238095238086

* neuroticism:
* random forest = 0.40956989247311826
* svm = 0.3704597701149425
* xgboost = 0.42496159754224266

* OutDoor:
* random forest = 0.711505376344086
* svm = 0.5251761216166111
* xgboost = 0.5456136447905079

# picture result

### Category: openness
1. xgboost : 0.39452838827838826
2. random tree : 0.4666666666666667
3. svm : 0.3719322344322344

### Category: conscientiousness

1. xgboost : 0.3487271062271063
2. random tree :0.3246199633699634
3. svm : 0.26666666666666666

### Category: extraversion

1. xgboost :0.35042582417582413
2. random tree :0.36567765567765564
3. svm : 0.39228021978021976

### Category: agreeableness

1. xgboost :0.41208333333333336
2. random tree : 0.35720238095238094
3. svm : 0.3710714285714286

### Category: neuroticism

1. xgboost :0.3020604395604395
2. random tree :0.33744505494505495
3. svm : 0.39228021978021976

### Category: outdoor

[[1 5]
[4 5]]
precision recall f1-score support

0 0.20 0.17 0.18 6
1 0.50 0.56 0.53 9

accuracy 0.40 15
macro avg 0.35 0.36 0.35 15
weighted avg 0.38 0.40 0.39 15

0.4
**average k-fold accuracy:** 0.5061904761904763

## Chi-square 500
### Category: openness

[[0 3 1]
[0 4 1]
[1 5 0]]
precision recall f1-score support

0 0.00 0.00 0.00 4
1 0.33 0.80 0.47 5
2 0.00 0.00 0.00 6

accuracy 0.27 15
macro avg 0.11 0.27 0.16 15
weighted avg 0.11 0.27 0.16 15

0.26666666666666666
0.3478296703296703

### Category: outdoor

[[4 2]
[0 9]]
precision recall f1-score support

0 1.00 0.67 0.80 6
1 0.82 1.00 0.90 9

accuracy 0.87 15
macro avg 0.91 0.83 0.85 15
weighted avg 0.89 0.87 0.86 15

0.8666666666666667
0.7642857142857142

## chi-square 50

### Category: openness

[[0 2 2]
[0 3 2]
[1 2 3]]
precision recall f1-score support

0 0.00 0.00 0.00 4
1 0.43 0.60 0.50 5
2 0.43 0.50 0.46 6

accuracy 0.40 15
macro avg 0.29 0.37 0.32 15
weighted avg 0.31 0.40 0.35 15

0.4
0.39645604395604395

### Category: outdoor

[[4 2]
[1 8]]
precision recall f1-score support

0 0.80 0.67 0.73 6
1 0.80 0.89 0.84 9

accuracy 0.80 15
macro avg 0.80 0.78 0.78 15
weighted avg 0.80 0.80 0.80 15

0.8
0.7104761904761905

# Summary:
tf-idf 4000:
1. openness
* Sample accuracy: 0.3333333333333333
* average k-fold accuracy: 0.3004945054945055
2. outdoor
* Sample accuracy: 0.4
* average k-fold accuracy: 0.5061904761904763

Chi-square 500:
1. openness
* Sample accuracy: 0.26666666666666666
* average k-fold accuracy: 0.3478296703296703
2. outdoor
* Sample accuracy: 0.8666666666666667
* average k-fold accuracy: 0.7642857142857142

Chi-square 50:
1. openness
* Sample accuracy: 0.4
* average k-fold accuracy: 0.39645604395604395
2. outdoor
* Sample accuracy: 0.8
* average k-fold accuracy: 0.7104761904761905

count Chi-square 500:
1. openness
* Sample accuracy: 0.3
* average k-fold accuracy: 0.3166666666666666
2. outdoor
* Sample accuracy: 0.5
* average k-fold accuracy: 0.576310715609937

count Chi-square 50:
1. openness
* Sample accuracy: 0.4
* average k-fold accuracy: 0.33689655172413796
2. outdoor
* Sample accuracy: 0.43333333333333335
* average k-fold accuracy: 0.5252836484983314

# Model difference(tf-idf chi 500)

## Personality
* Openess:
* random forest = 0.3466666666666667
* svm = 0.3704597701149425
* xgboost = 0.33999999999999997

* conscientiousness:
* random forest = 0.2949430584247047
* svm = 0.3704597701149425
* xgboost = 0.2691686530006886

* extraversion:
* random forest = 0.3579569892473119
* svm = 0.3704597701149425
* xgboost = 0.369715821812596

* aggreablness:
* random forest = 0.298963133640553
* svm = 0.3704597701149425
* xgboost = 0.30595238095238086

* neuroticism:
* random forest = 0.40956989247311826
* svm = 0.3704597701149425
* xgboost = 0.42496159754224266

* OutDoor:
* random forest = 0.711505376344086
* svm = 0.5251761216166111
* xgboost = 0.5456136447905079

把少的一邊複製，加重權重 smote
class weight
scale_pos_weight