Spaces:
Sleeping
Sleeping
Update deliverable2.py
Browse files- deliverable2.py +42 -43
deliverable2.py
CHANGED
@@ -1,6 +1,5 @@
|
|
1 |
import requests
|
2 |
from bs4 import BeautifulSoup
|
3 |
-
import pandas as pd
|
4 |
from sentence_transformers import SentenceTransformer, util
|
5 |
from transformers import pipeline
|
6 |
|
@@ -19,7 +18,7 @@ class URLValidator:
|
|
19 |
def fetch_page_content(self, url: str) -> str:
|
20 |
""" Fetches and extracts text content from the given URL, handling errors gracefully. """
|
21 |
try:
|
22 |
-
headers = {"User-Agent": "Mozilla/5.0"}
|
23 |
response = requests.get(url, timeout=10, headers=headers)
|
24 |
response.raise_for_status()
|
25 |
soup = BeautifulSoup(response.text, "html.parser")
|
@@ -81,50 +80,50 @@ class URLValidator:
|
|
81 |
return " ".join(reasons) if reasons else "This source is highly credible and relevant."
|
82 |
|
83 |
def rate_url_validity(self, user_query: str, url: str):
|
84 |
-
|
85 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
86 |
|
87 |
-
# If content fetching failed, return a properly structured response
|
88 |
-
if "Error" in content:
|
89 |
return {
|
90 |
-
"raw_score": {
|
91 |
-
"Domain Trust":
|
92 |
-
"Content Relevance":
|
93 |
-
"Fact-Check Score":
|
94 |
-
"Bias Score":
|
95 |
-
"Final Validity Score":
|
96 |
},
|
97 |
"stars": {
|
98 |
-
"icon":
|
99 |
},
|
100 |
-
"explanation":
|
101 |
}
|
102 |
-
|
103 |
-
domain_trust = self.get_domain_trust(url, content)
|
104 |
-
similarity_score = self.compute_similarity_score(user_query, content)
|
105 |
-
fact_check_score = self.check_facts(content)
|
106 |
-
bias_score = self.detect_bias(content)
|
107 |
-
|
108 |
-
final_score = (
|
109 |
-
(0.3 * domain_trust) +
|
110 |
-
(0.3 * similarity_score) +
|
111 |
-
(0.2 * fact_check_score) +
|
112 |
-
(0.2 * bias_score)
|
113 |
-
)
|
114 |
-
|
115 |
-
stars, icon = self.get_star_rating(final_score)
|
116 |
-
explanation = self.generate_explanation(domain_trust, similarity_score, fact_check_score, bias_score, final_score)
|
117 |
-
|
118 |
-
return {
|
119 |
-
"raw_score": {
|
120 |
-
"Domain Trust": domain_trust,
|
121 |
-
"Content Relevance": similarity_score,
|
122 |
-
"Fact-Check Score": fact_check_score,
|
123 |
-
"Bias Score": bias_score,
|
124 |
-
"Final Validity Score": final_score
|
125 |
-
},
|
126 |
-
"stars": {
|
127 |
-
"icon": icon
|
128 |
-
},
|
129 |
-
"explanation": explanation
|
130 |
-
}
|
|
|
1 |
import requests
|
2 |
from bs4 import BeautifulSoup
|
|
|
3 |
from sentence_transformers import SentenceTransformer, util
|
4 |
from transformers import pipeline
|
5 |
|
|
|
18 |
def fetch_page_content(self, url: str) -> str:
|
19 |
""" Fetches and extracts text content from the given URL, handling errors gracefully. """
|
20 |
try:
|
21 |
+
headers = {"User-Agent": "Mozilla/5.0"} # Helps bypass some bot protections
|
22 |
response = requests.get(url, timeout=10, headers=headers)
|
23 |
response.raise_for_status()
|
24 |
soup = BeautifulSoup(response.text, "html.parser")
|
|
|
80 |
return " ".join(reasons) if reasons else "This source is highly credible and relevant."
|
81 |
|
82 |
def rate_url_validity(self, user_query: str, url: str):
|
83 |
+
""" Main function to evaluate the validity of a webpage. """
|
84 |
+
content = self.fetch_page_content(url)
|
85 |
+
|
86 |
+
# If content fetching failed, return a properly structured response
|
87 |
+
if "Error" in content:
|
88 |
+
return {
|
89 |
+
"raw_score": {
|
90 |
+
"Domain Trust": 0,
|
91 |
+
"Content Relevance": 0,
|
92 |
+
"Fact-Check Score": 0,
|
93 |
+
"Bias Score": 0,
|
94 |
+
"Final Validity Score": 0
|
95 |
+
},
|
96 |
+
"stars": {
|
97 |
+
"icon": "❌"
|
98 |
+
},
|
99 |
+
"explanation": content # Display the error message
|
100 |
+
}
|
101 |
+
|
102 |
+
domain_trust = self.get_domain_trust(url, content)
|
103 |
+
similarity_score = self.compute_similarity_score(user_query, content)
|
104 |
+
fact_check_score = self.check_facts(content)
|
105 |
+
bias_score = self.detect_bias(content)
|
106 |
+
|
107 |
+
final_score = (
|
108 |
+
(0.3 * domain_trust) +
|
109 |
+
(0.3 * similarity_score) +
|
110 |
+
(0.2 * fact_check_score) +
|
111 |
+
(0.2 * bias_score)
|
112 |
+
)
|
113 |
+
|
114 |
+
stars, icon = self.get_star_rating(final_score)
|
115 |
+
explanation = self.generate_explanation(domain_trust, similarity_score, fact_check_score, bias_score, final_score)
|
116 |
|
|
|
|
|
117 |
return {
|
118 |
+
"raw_score": {
|
119 |
+
"Domain Trust": domain_trust,
|
120 |
+
"Content Relevance": similarity_score,
|
121 |
+
"Fact-Check Score": fact_check_score,
|
122 |
+
"Bias Score": bias_score,
|
123 |
+
"Final Validity Score": final_score
|
124 |
},
|
125 |
"stars": {
|
126 |
+
"icon": icon
|
127 |
},
|
128 |
+
"explanation": explanation
|
129 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|