Update README.md
Browse files
README.md
CHANGED
|
@@ -13,4 +13,63 @@ tags:
|
|
| 13 |
- sentiment
|
| 14 |
- turkish
|
| 15 |
- bert
|
| 16 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
- sentiment
|
| 14 |
- turkish
|
| 15 |
- bert
|
| 16 |
+
---
|
| 17 |
+
### Model Info
|
| 18 |
+
|
| 19 |
+
This model was developed/finetuned for movie review task for the Turkish Language. This model was finetuned via the Turkish movie review dataset.
|
| 20 |
+
- LABEL_0: positive review
|
| 21 |
+
- LABEL_1: negative review
|
| 22 |
+
|
| 23 |
+
### Model Sources
|
| 24 |
+
|
| 25 |
+
<!-- Provide the basic links for the model. -->
|
| 26 |
+
|
| 27 |
+
- **Dataset:** http://humirapps.cs.hacettepe.edu.tr/tsad.aspx
|
| 28 |
+
- **Paper:** https://dl.acm.org/doi/10.1145/3557892
|
| 29 |
+
- **Demo-Coding [optional]:** https://github.com/anil1055/Turkish_Sentiment_Analysis-Hotel-and-Movie-Reviews/tree/main
|
| 30 |
+
- **Finetuned from model [optional]:** https://huggingface.co/dbmdz/bert-base-turkish-uncased
|
| 31 |
+
|
| 32 |
+
#### Preprocessing
|
| 33 |
+
|
| 34 |
+
You must apply removing stopwords, stemming, or lemmatization process for Turkish.
|
| 35 |
+
|
| 36 |
+
### Results
|
| 37 |
+
|
| 38 |
+
- auprc = 0.9547155589592419
|
| 39 |
+
- auroc = 0.9567033960358541
|
| 40 |
+
- eval_loss = 0.4520341001172079
|
| 41 |
+
- fn = 1368
|
| 42 |
+
- fp = 1668
|
| 43 |
+
- mcc = 0.7727794159832003
|
| 44 |
+
- tn = 11682
|
| 45 |
+
- tp = 11982
|
| 46 |
+
- Accuracy: %92.11
|
| 47 |
+
|
| 48 |
+
## Citation
|
| 49 |
+
|
| 50 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 51 |
+
|
| 52 |
+
**BibTeX:**
|
| 53 |
+
|
| 54 |
+
*@article{10.1145/3557892,
|
| 55 |
+
author = {Guven, Zekeriya Anil},
|
| 56 |
+
title = {The Comparison of Language Models with a Novel Text Filtering Approach for Turkish Sentiment Analysis},
|
| 57 |
+
year = {2022},
|
| 58 |
+
issue_date = {February 2023},
|
| 59 |
+
publisher = {Association for Computing Machinery},
|
| 60 |
+
address = {New York, NY, USA},
|
| 61 |
+
volume = {22},
|
| 62 |
+
number = {2},
|
| 63 |
+
issn = {2375-4699},
|
| 64 |
+
url = {https://doi.org/10.1145/3557892},
|
| 65 |
+
doi = {10.1145/3557892},
|
| 66 |
+
journal = {ACM Trans. Asian Low-Resour. Lang. Inf. Process.},
|
| 67 |
+
month = {dec},
|
| 68 |
+
articleno = {55},
|
| 69 |
+
numpages = {16},
|
| 70 |
+
keywords = {Language model, sentiment analysis, social network, natural language processing, text classification, data analysis}
|
| 71 |
+
}*
|
| 72 |
+
|
| 73 |
+
**APA:**
|
| 74 |
+
|
| 75 |
+
*Guven, Z. A. (2022). The Comparison of Language Models with a Novel Text Filtering Approach for Turkish Sentiment Analysis. ACM Transactions on Asian and Low-Resource Language Information Processing, 22(2), 1-16.*
|