File size: 3,685 Bytes
c5e5728 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 |
import streamlit as st
import requests
from bs4 import BeautifulSoup
from transformers import pipeline
from sentence_transformers import SentenceTransformer, util
class URLValidator:
def __init__(self):
self.similarity_model = SentenceTransformer('sentence-transformers/all-mpnet-base-v2')
self.fake_news_classifier = pipeline("text-classification", model="mrm8488/bert-tiny-finetuned-fake-news-detection")
self.sentiment_analyzer = pipeline("text-classification", model="cardiffnlp/twitter-roberta-base-sentiment")
def fetch_page_content(self, url):
try:
response = requests.get(url, timeout=10)
response.raise_for_status()
soup = BeautifulSoup(response.text, "html.parser")
return " ".join([p.text for p in soup.find_all("p")])
except:
return ""
def compute_similarity_score(self, user_query, content):
if not content:
return 0
return int(util.pytorch_cos_sim(
self.similarity_model.encode(user_query),
self.similarity_model.encode(content)
).item() * 100)
def detect_bias(self, content):
if not content:
return 50
sentiment_result = self.sentiment_analyzer(content[:512])[0]
return 100 if sentiment_result["label"] == "POSITIVE" else 50 if sentiment_result["label"] == "NEUTRAL" else 30
def get_star_rating(self, score: float):
stars = max(1, min(5, round(score / 20)))
return stars, "⭐" * stars
def generate_explanation(self, domain_trust, similarity_score, fact_check_score, bias_score, citation_score, final_score):
reasons = []
if domain_trust < 50:
reasons.append("The source has low domain authority.")
if similarity_score < 50:
reasons.append("The content is not highly relevant to your query.")
if fact_check_score < 50:
reasons.append("Limited fact-checking verification found.")
if bias_score < 50:
reasons.append("Potential bias detected in the content.")
if citation_score < 30:
reasons.append("Few citations found for this content.")
return " ".join(reasons) if reasons else "This source is highly credible and relevant."
def rate_url_validity(self, user_query, url):
content = self.fetch_page_content(url)
similarity_score = self.compute_similarity_score(user_query, content)
bias_score = self.detect_bias(content)
domain_trust = 60 # Placeholder
fact_check_score = 70 # Placeholder
citation_score = 50 # Placeholder
final_score = (0.3 * domain_trust) + (0.3 * similarity_score) + (0.2 * fact_check_score) + (0.1 * bias_score) + (0.1 * citation_score)
stars, icon = self.get_star_rating(final_score)
explanation = self.generate_explanation(domain_trust, similarity_score, fact_check_score, bias_score, citation_score, final_score)
return {
"Final Score": f"{final_score:.2f}%",
"Star Rating": icon,
"Explanation": explanation
}
# Initialize validator
validator = URLValidator()
st.title("URL Credibility Checker")
url = st.text_input("Enter URL")
query = st.text_input("Enter Search Query")
if st.button("Check Credibility"):
if url and query:
result = validator.rate_url_validity(query, url)
st.write("**Final Score:**", result["Final Score"])
st.write("**Star Rating:**", result["Star Rating"])
st.write("**Explanation:**", result["Explanation"])
else:
st.warning("Please enter both a URL and a search query.")
|