DexterSptizu
commited on
Commit
•
a0d2064
1
Parent(s):
986a219
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,136 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import spacy
|
3 |
+
from collections import Counter
|
4 |
+
from string import punctuation
|
5 |
+
|
6 |
+
# Load the English language model
|
7 |
+
nlp = spacy.load("en_core_web_sm")
|
8 |
+
|
9 |
+
# Example texts
|
10 |
+
EXAMPLES = {
|
11 |
+
"Scientific Abstract": """
|
12 |
+
Compatibility of systems of linear constraints over the set of natural numbers.
|
13 |
+
Criteria of compatibility of a system of linear Diophantine equations, strict inequations,
|
14 |
+
and nonstrict inequations are considered. Upper bounds for components of a minimal set of solutions
|
15 |
+
and algorithms of construction of minimal generating sets of solutions for all types of systems are given.
|
16 |
+
""",
|
17 |
+
"News Article": """
|
18 |
+
Machine learning is revolutionizing the way we interact with technology.
|
19 |
+
Artificial intelligence systems are becoming more sophisticated, enabling automated decision making
|
20 |
+
and pattern recognition at unprecedented scales. Deep learning algorithms continue to improve,
|
21 |
+
making breakthroughs in natural language processing and computer vision.
|
22 |
+
""",
|
23 |
+
"Technical Documentation": """
|
24 |
+
The user interface provides intuitive navigation through contextual menus and adaptive layouts.
|
25 |
+
System responses are optimized for performance while maintaining high reliability standards.
|
26 |
+
Database connections are pooled to minimize resource overhead and maximize throughput.
|
27 |
+
"""
|
28 |
+
}
|
29 |
+
|
30 |
+
def extract_keywords(text, num_keywords, extraction_type, include_phrases):
|
31 |
+
doc = nlp(text)
|
32 |
+
|
33 |
+
# Remove stopwords and punctuation
|
34 |
+
words = [token.text.lower() for token in doc
|
35 |
+
if not token.is_stop and not token.is_punct and token.text.strip()]
|
36 |
+
|
37 |
+
# Extract noun phrases if requested
|
38 |
+
phrases = []
|
39 |
+
if include_phrases:
|
40 |
+
phrases = [chunk.text.lower() for chunk in doc.noun_chunks
|
41 |
+
if len(chunk.text.split()) > 1]
|
42 |
+
|
43 |
+
# Extract keywords based on selected method
|
44 |
+
keywords = []
|
45 |
+
if extraction_type == "Nouns":
|
46 |
+
keywords = [token.text.lower() for token in doc
|
47 |
+
if token.pos_ == "NOUN" and not token.is_stop]
|
48 |
+
elif extraction_type == "Named Entities":
|
49 |
+
keywords = [ent.text.lower() for ent in doc.ents]
|
50 |
+
elif extraction_type == "All Words":
|
51 |
+
keywords = words
|
52 |
+
|
53 |
+
# Combine keywords and phrases
|
54 |
+
all_keywords = keywords + phrases
|
55 |
+
|
56 |
+
# Count frequencies
|
57 |
+
keyword_freq = Counter(all_keywords)
|
58 |
+
|
59 |
+
# Sort by frequency and get top keywords
|
60 |
+
top_keywords = sorted(keyword_freq.items(), key=lambda x: x[1], reverse=True)[:num_keywords]
|
61 |
+
|
62 |
+
# Format output
|
63 |
+
result = []
|
64 |
+
for idx, (keyword, freq) in enumerate(top_keywords, 1):
|
65 |
+
result.append(f"{idx}. {keyword} (frequency: {freq})")
|
66 |
+
|
67 |
+
return "\n".join(result) if result else "No keywords found."
|
68 |
+
|
69 |
+
def load_example(example_name):
|
70 |
+
return EXAMPLES.get(example_name, "")
|
71 |
+
|
72 |
+
# Create Gradio interface
|
73 |
+
with gr.Blocks(title="Keyword Extraction Tool") as demo:
|
74 |
+
gr.Markdown("# 🔍 Advanced NLP Keyword Extraction")
|
75 |
+
gr.Markdown("Extract keywords using spaCy's natural language processing")
|
76 |
+
|
77 |
+
with gr.Row():
|
78 |
+
with gr.Column(scale=2):
|
79 |
+
input_text = gr.Textbox(
|
80 |
+
label="Input Text",
|
81 |
+
placeholder="Enter your text here...",
|
82 |
+
lines=8
|
83 |
+
)
|
84 |
+
example_dropdown = gr.Dropdown(
|
85 |
+
choices=list(EXAMPLES.keys()),
|
86 |
+
label="Load Example Text"
|
87 |
+
)
|
88 |
+
|
89 |
+
with gr.Column(scale=1):
|
90 |
+
extraction_type = gr.Radio(
|
91 |
+
choices=["Nouns", "Named Entities", "All Words"],
|
92 |
+
value="Nouns",
|
93 |
+
label="Extraction Method"
|
94 |
+
)
|
95 |
+
|
96 |
+
include_phrases = gr.Checkbox(
|
97 |
+
label="Include Noun Phrases",
|
98 |
+
value=True
|
99 |
+
)
|
100 |
+
|
101 |
+
num_keywords = gr.Slider(
|
102 |
+
minimum=1,
|
103 |
+
maximum=20,
|
104 |
+
value=10,
|
105 |
+
step=1,
|
106 |
+
label="Number of Keywords"
|
107 |
+
)
|
108 |
+
|
109 |
+
extract_btn = gr.Button("Extract Keywords", variant="primary")
|
110 |
+
|
111 |
+
output_text = gr.Textbox(
|
112 |
+
label="Extracted Keywords",
|
113 |
+
lines=10,
|
114 |
+
interactive=False
|
115 |
+
)
|
116 |
+
|
117 |
+
# Set up event handlers
|
118 |
+
example_dropdown.change(
|
119 |
+
load_example,
|
120 |
+
inputs=[example_dropdown],
|
121 |
+
outputs=[input_text]
|
122 |
+
)
|
123 |
+
|
124 |
+
extract_btn.click(
|
125 |
+
extract_keywords,
|
126 |
+
inputs=[
|
127 |
+
input_text,
|
128 |
+
num_keywords,
|
129 |
+
extraction_type,
|
130 |
+
include_phrases
|
131 |
+
],
|
132 |
+
outputs=[output_text]
|
133 |
+
)
|
134 |
+
|
135 |
+
# Launch the app
|
136 |
+
demo.launch()
|