added README
Browse files
README.md
CHANGED
@@ -9,36 +9,37 @@ metrics:
|
|
9 |
model-index:
|
10 |
- name: sentiment-analysis-browser-extension
|
11 |
results: []
|
|
|
|
|
12 |
---
|
13 |
|
14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
should probably proofread and complete it, then remove this comment. -->
|
16 |
|
17 |
-
#
|
|
|
18 |
|
19 |
-
|
20 |
-
It achieves the following results on the evaluation set:
|
21 |
-
- Loss: 0.4768
|
22 |
-
- Accuracy: 0.8615
|
23 |
-
- F1: 0.8816
|
24 |
|
25 |
-
|
26 |
|
27 |
-
|
|
|
28 |
|
29 |
## Intended uses & limitations
|
|
|
30 |
|
31 |
-
|
32 |
-
|
33 |
-
## Training and evaluation data
|
34 |
|
35 |
-
|
|
|
|
|
|
|
36 |
|
37 |
## Training procedure
|
38 |
|
39 |
-
|
40 |
|
41 |
-
The following hyperparameters were used during training:
|
42 |
- learning_rate: 2e-05
|
43 |
- train_batch_size: 16
|
44 |
- eval_batch_size: 16
|
@@ -47,10 +48,6 @@ The following hyperparameters were used during training:
|
|
47 |
- lr_scheduler_type: linear
|
48 |
- num_epochs: 6
|
49 |
|
50 |
-
### Training results
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
### Framework versions
|
55 |
|
56 |
- Transformers 4.34.0
|
|
|
9 |
model-index:
|
10 |
- name: sentiment-analysis-browser-extension
|
11 |
results: []
|
12 |
+
language:
|
13 |
+
- en
|
14 |
---
|
15 |
|
16 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
should probably proofread and complete it, then remove this comment. -->
|
18 |
|
19 |
+
# Fine-tuned BERT model
|
20 |
+
We open source this fine-tuned BERT model to identify critical aspects within user reviews of adblocking extensions. For every user review, the model provides a criticality score (in the range of -1 to 1) with the negative scores signifying higher probability of finding critical topics within in the reviews.
|
21 |
|
22 |
+
We have used the [`distilbert-base-uncased`](https://huggingface.co/distilbert-base-uncased) as the base model and fine-tuned it on a manually annotated dataset of webstore reviews.
|
|
|
|
|
|
|
|
|
23 |
|
24 |
+
Further details can be found in our AsiaCCS paper - [From User Insights to Actionable Metrics: A User-Focused Evaluation of Privacy-Preserving Browser Extensions](https://doi.org/10.1145/3634737.3657028).
|
25 |
|
26 |
+
### Note
|
27 |
+
We haven't tested its accuracy on user reviews from other categories but are open to discuss the possibility of extrapolating it to other product categories.
|
28 |
|
29 |
## Intended uses & limitations
|
30 |
+
The model has been released to be freely used. It has not been trained on any private user data. Please do cite the above paper in our works.
|
31 |
|
32 |
+
## Evaluation data
|
|
|
|
|
33 |
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.4768
|
36 |
+
- Accuracy: 0.8615
|
37 |
+
- F1: 0.8816
|
38 |
|
39 |
## Training procedure
|
40 |
|
41 |
+
The training dataset comprised on 620 reviews and the test dataset had 150 reviews. The following hyperparameters were used during training:
|
42 |
|
|
|
43 |
- learning_rate: 2e-05
|
44 |
- train_batch_size: 16
|
45 |
- eval_batch_size: 16
|
|
|
48 |
- lr_scheduler_type: linear
|
49 |
- num_epochs: 6
|
50 |
|
|
|
|
|
|
|
|
|
51 |
### Framework versions
|
52 |
|
53 |
- Transformers 4.34.0
|