Matcha Sentiment
Examples
{
"input_files": [
"0": "Matchaya Gandaria City + IKUYO (done; need recheck).xlsx" ,
"1": "Feel Matcha Braga - Sementara.xlsx"
] ,
"output_file": "D:/matcha sentiment/data/processed/matcha_sentiment_binary.csv" ,
"original_rows": 2392 ,
"dropped_netral": 14 ,
"dropped_other_label": 0 ,
"dropped_bad_text": 0 ,
"dropped_duplicates": 219 ,
"balanced": true ,
"random_state": 42 ,
"kept_rows_before_balance": 2159 ,
"labels_before_balance": {
"Negatif": 1145 ,
"Positif": 1014
} ,
"balance_target_per_label": 1014 ,
"dropped_by_balance": 131 ,
"kept_rows": 2028 ,
"labels": {
"Negatif": 1014 ,
"Positif": 1014
}
}
TF-IDF dan Word2Vec 10-fold
accuracy | precision | recall | f1 | roc_auc | feature | model | folds | n |
|---|---|---|---|---|---|---|---|---|
0.9684418145956608 | 0.9788306451612904 | 0.9575936883629192 | 0.9680957128614156 | 0.9951448945531786 | word2vec | logistic_regression | 10 | 2028 |
Transformer
model_id | slug | test_accuracy | test_precision | test_recall | test_f1 | test_roc_auc |
|---|---|---|---|---|---|---|
sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 | sentence-transformers__paraphrase-multilingual-MiniLM-L12-v2 | 0.9950738916256158 | 0.9901960784313726 | 0.99009900990099 | 0.9950738916256158 | 0.9998058629392352 |
Top words TF-IDF
term | weight | label_name |
|---|---|---|
sangat direkomendasikan | -3.7154029272590687 | Positif |
Keyword penting
term | positif_docs | negatif_docs | positif_rate | negatif_rate | dominant_label | lift |
|---|---|---|---|---|---|---|
direkomendasikan | 510 | 380 | 0.0483234714003944 | 0.0009861932938856 | Positif | 2.8166665873305582e-8 |