Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,32 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
language: en
|
4 |
+
tags:
|
5 |
+
- sentence similarity
|
6 |
+
- semantic search
|
7 |
+
widget:
|
8 |
+
- text:
|
9 |
+
- I like batman
|
10 |
+
- I watch super hero movies
|
11 |
+
pipeline_tag: sentence-similarity
|
12 |
---
|
13 |
+
|
14 |
+
# Dataset Collection:
|
15 |
+
* The news dataset is collected from Kaggle.
|
16 |
+
* The dataset has news title ,news content and the label(the label shows the cosine similarity between news title and news content).
|
17 |
+
* Different strategies have been followed during the data gathering phase.
|
18 |
+
|
19 |
+
# sentence transformer is fine-tuned for semantic search and sentence similarity
|
20 |
+
* The model is fine-tuned on the dataset.
|
21 |
+
* This model can be used for semantic search,sentence similarity,recommendation system.
|
22 |
+
* This model can be used for the inference purpose as well.
|
23 |
+
|
24 |
+
# Data Fields:
|
25 |
+
|
26 |
+
**label**: cosine similarity between news title and news content
|
27 |
+
**news title**: The title of the news
|
28 |
+
**news content**:The content of the news
|
29 |
+
|
30 |
+
# Application:
|
31 |
+
* This model is useful for the semantic search,sentence similarity,recommendation system.
|
32 |
+
* You can fine-tune this model for your particular use cases.
|