Spaces:
Sleeping
Sleeping
Updated README
Browse files
README.md
CHANGED
@@ -8,26 +8,27 @@ pinned: false
|
|
8 |
python_version: 3.10.5
|
9 |
---
|
10 |
|
11 |
-
|
12 |
|
13 |
-
|
14 |
|
15 |
-
|
16 |
|
17 |
-
|
18 |
|
19 |
-
|
20 |
|
21 |
-
|
|
|
22 |
|
23 |
-
|
24 |
|
25 |
-
|
26 |
-
|
27 |
-
-
|
28 |
-
-
|
29 |
-
|
30 |
-
|
31 |
|
32 |
### Aggression Detection Results
|
33 |
|
@@ -41,7 +42,7 @@ This app detects presence of Aggression and Misogyny in online social media text
|
|
41 |
| -------- | ----- |
|
42 |
| F1 Score | 0.852 |
|
43 |
|
44 |
-
## How to Run
|
45 |
|
46 |
<!-- Installation and Running Steps -->
|
47 |
|
@@ -51,6 +52,6 @@ This app detects presence of Aggression and Misogyny in online social media text
|
|
51 |
|
52 |
## Additional Links
|
53 |
|
54 |
-
|
55 |
-
|
56 |
- [[PDF] Paper published at ICON 2021](https://aclanthology.org/2021.icon-main.60.pdf)
|
|
|
|
8 |
python_version: 3.10.5
|
9 |
---
|
10 |
|
11 |
+
## Agression and Misogyny Detection App
|
12 |
|
13 |
+
Social media platforms have become hotspots for the proliferation of trolling, aggression, and hate speech. With an overwhelming volume of social media data being generated every day, manual inspection is simply impractical. In response to this pressing issue, we present an efficient and rapid method for detecting aggression and misogyny in online social media texts.
|
14 |
|
15 |
+
What sets our model apart is not only its high performance but also its significantly reduced training time, model size, and resource requirements. These advantages make our model highly practical for fast inference, ensuring prompt identification of aggression and misogyny in online social media texts.
|
16 |
|
17 |
+
> [Try it out here](https://huggingface.co/spaces/sdutta28/AggDetectApp)
|
18 |
|
19 |
+
### Features
|
20 |
|
21 |
+
- Detection of Aggression and Misogyny in texts
|
22 |
+
- LIME based prediction for explainability
|
23 |
|
24 |
+
### Tech Stack
|
25 |
|
26 |
+
- Python
|
27 |
+
- XgBoost
|
28 |
+
- Scikit-Learn
|
29 |
+
- HuggingFace Transformers
|
30 |
+
- LIME
|
31 |
+
- Docker
|
32 |
|
33 |
### Aggression Detection Results
|
34 |
|
|
|
42 |
| -------- | ----- |
|
43 |
| F1 Score | 0.852 |
|
44 |
|
45 |
+
## How to Run Locally
|
46 |
|
47 |
<!-- Installation and Running Steps -->
|
48 |
|
|
|
52 |
|
53 |
## Additional Links
|
54 |
|
55 |
+
- [Try the App Here](https://huggingface.co/spaces/sdutta28/AggDetectApp)
|
|
|
56 |
- [[PDF] Paper published at ICON 2021](https://aclanthology.org/2021.icon-main.60.pdf)
|
57 |
+
- [Model training Repo](https://github.com/Dutta-SD/AggDetect)
|