sdutta28 commited on
Commit
1198436
1 Parent(s): 15c875a

Updated README

Browse files
Files changed (1) hide show
  1. README.md +17 -16
README.md CHANGED
@@ -8,26 +8,27 @@ pinned: false
8
  python_version: 3.10.5
9
  ---
10
 
11
- # Agression and Misogyny Detection App
12
 
13
- <!-- Intro and about the project -->
14
 
15
- This app detects presence of Aggression and Misogyny in online social media texts. Try it out [here](https://huggingface.co/spaces/sdutta28/AggDetectApp)
16
 
17
- ## Technologies Used
18
 
19
- <!-- Tech stack, libraries etc -->
20
 
21
- `Python, Transformers, XgBoost, Scikit-Learn`
 
22
 
23
- ## Workflow
24
 
25
- <!-- In some detail of how this works -->
26
-
27
- - A text is given to the app
28
- - It predicts aggression and misogyny in the application
29
-
30
- ## Results
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
- <!-- Kaggle model training links -->
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)