Sumayyea commited on
Commit
33cc687
Β·
verified Β·
1 Parent(s): 00b8a84

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +44 -16
app.py CHANGED
@@ -1,7 +1,7 @@
1
  import streamlit as st
2
  import re
3
  from datetime import datetime
4
- from typing import Dict, Any, Optional
5
 
6
 
7
  # -------------------------
@@ -9,6 +9,13 @@ from typing import Dict, Any, Optional
9
  # -------------------------
10
 
11
  class DocumentClassifier:
 
 
 
 
 
 
 
12
  def __init__(self):
13
  self.categories = {
14
  'ADMINISTRATIVE': ['memo', 'circular', 'notification', 'order', 'policy'],
@@ -17,24 +24,27 @@ class DocumentClassifier:
17
  'HR': ['recruitment', 'promotion', 'transfer', 'leave', 'employee'],
18
  'TECHNICAL': ['specification', 'tender', 'project', 'development', 'infrastructure']
19
  }
 
 
20
 
21
  def extract_features(self, document_text: str) -> Dict[str, Any]:
 
22
  text_lower = document_text.lower()
23
  features = {}
24
 
25
- # Score each category based on keywords
26
  for category, keywords in self.categories.items():
27
  feature_count = sum(text_lower.count(keyword) for keyword in keywords)
28
  features[f'{category}_score'] = feature_count
29
 
30
  features['word_count'] = len(text_lower.split())
31
 
32
- # Date patterns (extract actual date if found)
33
  date_patterns = [
34
- r'\b\d{1,2}[-/]\d{1,2}[-/]\d{4}\b', # 12/02/2023
35
- r'\b\d{1,2}(?:st|nd|rd|th)?\s+[A-Z][a-z]+,\s+\d{4}\b', # 7th January, 2025
36
- r'\b\d{1,2}(?:st|nd|rd|th)?\s+[A-Z][a-z]+\s+\d{4}\b', # 29 January 2025
37
- r'\b[A-Z][a-z]+\s+\d{1,2},\s+\d{4}\b' # January 29, 2025
38
  ]
39
  found_date = None
40
  for pattern in date_patterns:
@@ -45,10 +55,10 @@ class DocumentClassifier:
45
  features['has_date'] = 1 if found_date else 0
46
  features['extracted_date'] = found_date if found_date else "N/A"
47
 
48
- # Reference number patterns (extract actual reference if found)
49
  ref_patterns = [
50
- r'\bref(?:\.|\s|:)*[A-Za-z0-9/\-()]+', # Ref No: XYZ-123
51
- r'\bno(?:\.|\s|:)*[A-Za-z0-9/\-()]+', # No. DD/XYZ-123
52
  ]
53
  found_ref = None
54
  for pattern in ref_patterns:
@@ -62,11 +72,13 @@ class DocumentClassifier:
62
  return features
63
 
64
  def classify_document(self, document_text: str, filename: str = "") -> Dict[str, Any]:
 
65
  features = self.extract_features(document_text)
66
  scores = {}
67
 
68
  for category in self.categories:
69
- base_score = features.get(f'{category}_score', 0)
 
70
  normalized_score = base_score / max(features['word_count'] / 100, 1)
71
  scores[category] = normalized_score
72
 
@@ -82,12 +94,21 @@ class DocumentClassifier:
82
  'features': features
83
  }
84
 
 
 
 
 
 
 
 
 
 
85
 
86
  # -------------------------
87
- # Streamlit UI
88
  # -------------------------
89
 
90
- st.set_page_config(page_title="Document Classifier", layout="centered")
91
  st.title("πŸ“„ AI-Based Document Classifier")
92
  st.subheader("Upload government text files for automatic classification")
93
 
@@ -97,17 +118,24 @@ uploaded_files = st.file_uploader(
97
 
98
  if uploaded_files:
99
  classifier = DocumentClassifier()
 
 
 
100
  for uploaded_file in uploaded_files:
101
  filename = uploaded_file.name
102
  content = uploaded_file.read().decode("utf-8")
103
- result = classifier.classify_document(content, filename)
 
 
104
 
 
 
105
  st.markdown("---")
106
  st.markdown(f"πŸ“ **Document:** `{result['filename']}`")
107
  st.markdown(f"πŸ“‚ **Predicted Category:** `{result['predicted_category']}`")
108
  st.markdown(f"πŸ“Š **Confidence:** `{result['confidence']:.2f}`")
109
 
110
- # Show extracted date and reference if found
111
  extracted_date = result["features"].get("extracted_date", "N/A")
112
  extracted_ref = result["features"].get("extracted_reference", "N/A")
113
 
@@ -121,4 +149,4 @@ if uploaded_files:
121
  st.json(result['features'])
122
 
123
  with st.expander("πŸ“œ Full Text", expanded=False):
124
- st.text(content)
 
1
  import streamlit as st
2
  import re
3
  from datetime import datetime
4
+ from typing import Dict, Any, List
5
 
6
 
7
  # -------------------------
 
9
  # -------------------------
10
 
11
  class DocumentClassifier:
12
+ """
13
+ AI Document Classification System for Government E-Filing
14
+
15
+ This class demonstrates how AI can automatically classify government documents
16
+ into predefined categories based on content analysis.
17
+ """
18
+
19
  def __init__(self):
20
  self.categories = {
21
  'ADMINISTRATIVE': ['memo', 'circular', 'notification', 'order', 'policy'],
 
24
  'HR': ['recruitment', 'promotion', 'transfer', 'leave', 'employee'],
25
  'TECHNICAL': ['specification', 'tender', 'project', 'development', 'infrastructure']
26
  }
27
+ self.confidence_threshold = 0.7
28
+ print("Document Classifier initialized with government document categories")
29
 
30
  def extract_features(self, document_text: str) -> Dict[str, Any]:
31
+ """Extract features from document text for classification"""
32
  text_lower = document_text.lower()
33
  features = {}
34
 
35
+ # Count occurrences of category-specific keywords
36
  for category, keywords in self.categories.items():
37
  feature_count = sum(text_lower.count(keyword) for keyword in keywords)
38
  features[f'{category}_score'] = feature_count
39
 
40
  features['word_count'] = len(text_lower.split())
41
 
42
+ # Date extraction
43
  date_patterns = [
44
+ r'\b\d{1,2}[-/]\d{1,2}[-/]\d{4}\b',
45
+ r'\b\d{1,2}(?:st|nd|rd|th)?\s+[A-Z][a-z]+,\s+\d{4}\b',
46
+ r'\b\d{1,2}(?:st|nd|rd|th)?\s+[A-Z][a-z]+\s+\d{4}\b',
47
+ r'\b[A-Z][a-z]+\s+\d{1,2},\s+\d{4}\b'
48
  ]
49
  found_date = None
50
  for pattern in date_patterns:
 
55
  features['has_date'] = 1 if found_date else 0
56
  features['extracted_date'] = found_date if found_date else "N/A"
57
 
58
+ # Reference number extraction
59
  ref_patterns = [
60
+ r'\bref(?:\.|\s|:)*[A-Za-z0-9/\-()]+',
61
+ r'\bno(?:\.|\s|:)*[A-Za-z0-9/\-()]+'
62
  ]
63
  found_ref = None
64
  for pattern in ref_patterns:
 
72
  return features
73
 
74
  def classify_document(self, document_text: str, filename: str = "") -> Dict[str, Any]:
75
+ """Classify a document using AI-based content analysis"""
76
  features = self.extract_features(document_text)
77
  scores = {}
78
 
79
  for category in self.categories:
80
+ score_key = f'{category}_score'
81
+ base_score = features.get(score_key, 0)
82
  normalized_score = base_score / max(features['word_count'] / 100, 1)
83
  scores[category] = normalized_score
84
 
 
94
  'features': features
95
  }
96
 
97
+ def batch_classify(self, documents: List[Dict[str, str]]) -> List[Dict[str, Any]]:
98
+ """Classify multiple documents in batch"""
99
+ results = []
100
+ for doc in documents:
101
+ result = self.classify_document(doc['content'], doc.get('filename', ''))
102
+ results.append(result)
103
+ print(f"Batch classification completed for {len(documents)} documents")
104
+ return results
105
+
106
 
107
  # -------------------------
108
+ # Streamlit App UI
109
  # -------------------------
110
 
111
+ st.set_page_config(page_title="AI Document Classifier", layout="centered")
112
  st.title("πŸ“„ AI-Based Document Classifier")
113
  st.subheader("Upload government text files for automatic classification")
114
 
 
118
 
119
  if uploaded_files:
120
  classifier = DocumentClassifier()
121
+
122
+ # Read all files for batch processing
123
+ docs = []
124
  for uploaded_file in uploaded_files:
125
  filename = uploaded_file.name
126
  content = uploaded_file.read().decode("utf-8")
127
+ docs.append({"filename": filename, "content": content})
128
+
129
+ results = classifier.batch_classify(docs)
130
 
131
+ # Display results per document
132
+ for result in results:
133
  st.markdown("---")
134
  st.markdown(f"πŸ“ **Document:** `{result['filename']}`")
135
  st.markdown(f"πŸ“‚ **Predicted Category:** `{result['predicted_category']}`")
136
  st.markdown(f"πŸ“Š **Confidence:** `{result['confidence']:.2f}`")
137
 
138
+ # Extracted metadata
139
  extracted_date = result["features"].get("extracted_date", "N/A")
140
  extracted_ref = result["features"].get("extracted_reference", "N/A")
141
 
 
149
  st.json(result['features'])
150
 
151
  with st.expander("πŸ“œ Full Text", expanded=False):
152
+ st.text(next(doc['content'] for doc in docs if doc['filename'] == result['filename']))