README.md

---
license: mit
language:
- en
base_model:
- google-bert/bert-base-uncased
pipeline_tag: text-classification
---
# News Relevancy Classifiers

## bert-ft-v2

![BERTft Badge](https://img.shields.io/badge/Model-BERT--ft--v2-blue)

### Model Description
- **Purpose**: This model is trained for a specific task in research, it is not a commmercial product and should not be used in for-profit.
- **Architecture**: `bert-base-uncased`  
- **Fine-tuning task**: Four-class English healthcare and AI news-headline relevancy classification  
- **Dataset**: ~254 English headlines (2024–2025) manually labeled into:
  - 0 — Not Relevant  
  - 1 — Least Relevant  
  - 2 — Highly Relevant  
  - 3 — Most Relevant  
- **HF Repo**: [`cloud0day3/bert-ft-v2`](https://huggingface.co/cloud0day3/bert-ft-v2) (latest v3 checkpoint, 6 June 2025)  
- **Date Trained**: 2025-06-06  

#### Model Inputs

- A raw English headline (string), truncated/padded to 96 tokens.  
- Tokenization handled by the bundled `vocab.txt` + `tokenizer_config.json` + `special_tokens_map.json`.

#### Model Outputs

- A single integer label (0–3). Mapped to human-readable categories:
  ```python
  LABELS = {
      0: "Not Relevant",
      1: "Least Relevant",
      2: "Highly Relevant",
      3: "Most Relevant"
  }


#### Intended Use
- **Primary**: Automatically assign a relevancy score to healthcare and AI English news headlines so that downstream pipelines (e.g., filtering, ranking) can operate without manual triage.

#### Examples of use:

- Pre-filtering a news aggregation feed to capture healthcare and AI news.

- Prioritizing headlines for editorial review.

- Input to summarization/retrieval pipelines.

#### Out-of-Scope Uses
- Any non-English text.

- Multi-sentence inputs or full articles (this model is tuned on single-sentence headlines).

- Tasks other than healthcare-tech relevancy (e.g., sentiment analysis, topic modeling).

- High-risk decision making without human oversight (e.g., emergency alerts).