Update README.md

README.md

@@ -1,50 +1,52 @@
 ---
 library_name: transformers
-tags: []
+tags:
+  - text-classification
+  - malicious-url-detection
 ---
 
 # Malicious-Url-Detection
 
-Leveraging this model, you can identify harmful links intended to exploit users—such as phishing or malware URLs—by accurately classifying them as either malicious or benign.
+Leveraging this fine-tuned model, you can identify harmful links intended to exploit users—such as phishing or malware URLs—by accurately classifying them as either malicious or benign.
 
 ## Model Details
 
 ### Model Description
 
-This is the model card for a Hugging Face transformers model designed specifically for malicious URL detection. It has been fine-tuned to classify URLs as either malicious or benign by leveraging deep learning techniques that capture subtle textual patterns indicative of phishing or malware.
+This model is a **fine-tuned** version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased), adapted specifically for malicious URL detection. It employs a text-classification approach to distinguish between benign and malicious URLs. By learning patterns from a curated dataset of phishing, malware, and legitimate URLs, it aims to help users and organizations bolster their defenses against a range of cyber threats.
 
-- **Developed by:** Eason Liu
-- **Language(s):** English
+- **Developed by:** Eason Liu  
+- **Language:** English  
+- **Model Type:** Text Classification (URL-focused)  
+- **Finetuned From:** [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased)
 
-## Uses
+## Intended Use
 
 ### Direct Use
-
-This model can be directly employed via the Hugging Face transformers pipeline to classify URLs. It is intended for use in applications that require the identification of phishing and malware URLs, helping users or security systems to flag and filter potentially harmful links in real-time.
+- **URL Classification:** Detect whether a URL is malicious (e.g., phishing, malware) or benign.
+- **Security Pipelines:** Integrate into email filtering systems or website scanning tools to flag harmful links.
 
 ### Out-of-Scope Use
+- General text classification tasks not related to malicious URL detection.
+- Tasks requiring more nuanced context beyond the URL string (e.g., domain reputation, real-time link behavior).
 
-This model is specialized for malicious URL detection. It is not designed for tasks such as general text classification, sentiment analysis, or language translation.
-
-## Bias, Risks, and Limitations
-
-Users should note that the model’s performance depends on the representativeness of its training data. It may reflect biases inherent in the dataset and could produce false positives or negatives in certain scenarios. Therefore, it is recommended to incorporate this model as part of a broader cybersecurity framework rather than relying on it as a standalone solution.
-
-### Recommendations
+## How to Get Started
 
-To mitigate risks and improve reliability, users are encouraged to regularly update the training data to account for evolving cyber threats. Additionally, integrating this model with complementary security measures can help balance its limitations and enhance overall threat detection.
-
-## How to Get Started with the Model
-
-To begin using this model, install the Hugging Face transformers library and load the model with the following code snippet:
+Below is a quick example showing how to use this model with the 🤗 Transformers `pipeline`:
 
 ```python
 from transformers import pipeline
 
-# Load the malicious URL detection model
-classifier = pipeline("text-classification", model="CrabInHoney/urlbert-tiny-v3-malicious-url-classifier")
+# Initialize the text-classification pipeline with this fine-tuned model
+classifier = pipeline(
+    "text-classification",
+    model="Eason918/malicious-url-detector",
+    truncation=True
+)
 
-# Example usage
+# Example URL
 url = "http://example.com/suspicious-link"
+
+# Classify the URL
 result = classifier(url)
 print(result)