Upload app.py
Browse files
app.py
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import streamlit as st
|
| 2 |
+
import bs4 as bs
|
| 3 |
+
import urllib.request
|
| 4 |
+
import re
|
| 5 |
+
import nltk
|
| 6 |
+
import heapq
|
| 7 |
+
|
| 8 |
+
# NLTK downloads
|
| 9 |
+
nltk.download('punkt')
|
| 10 |
+
nltk.download('stopwords')
|
| 11 |
+
|
| 12 |
+
# Function to summarize text
|
| 13 |
+
def summarize_text(url):
|
| 14 |
+
scraped_data = urllib.request.urlopen(url)
|
| 15 |
+
article = scraped_data.read()
|
| 16 |
+
parsed_article = bs.BeautifulSoup(article,'lxml')
|
| 17 |
+
paragraphs = parsed_article.find_all('p')
|
| 18 |
+
article_text = ""
|
| 19 |
+
for p in paragraphs:
|
| 20 |
+
article_text += p.text
|
| 21 |
+
article_text = re.sub(r'\[[0-9]*\]', ' ', article_text)
|
| 22 |
+
article_text = re.sub(r'\s+', ' ', article_text)
|
| 23 |
+
formatted_article_text = re.sub('[^a-zA-Z]', ' ', article_text )
|
| 24 |
+
formatted_article_text = re.sub(r'\s+', ' ', formatted_article_text)
|
| 25 |
+
sentence_list = nltk.sent_tokenize(article_text)
|
| 26 |
+
stopwords = nltk.corpus.stopwords.words('english')
|
| 27 |
+
word_frequencies = {}
|
| 28 |
+
for word in nltk.word_tokenize(formatted_article_text):
|
| 29 |
+
if word not in stopwords:
|
| 30 |
+
if word not in word_frequencies.keys():
|
| 31 |
+
word_frequencies[word] = 1
|
| 32 |
+
else:
|
| 33 |
+
word_frequencies[word] += 1
|
| 34 |
+
maximum_frequncy = max(word_frequencies.values())
|
| 35 |
+
for word in word_frequencies.keys():
|
| 36 |
+
word_frequencies[word] = (word_frequencies[word]/maximum_frequncy)
|
| 37 |
+
sentence_scores = {}
|
| 38 |
+
for sent in sentence_list:
|
| 39 |
+
for word in nltk.word_tokenize(sent.lower()):
|
| 40 |
+
if word in word_frequencies.keys():
|
| 41 |
+
if len(sent.split(' ')) < 30:
|
| 42 |
+
if sent not in sentence_scores.keys():
|
| 43 |
+
sentence_scores[sent] = word_frequencies[word]
|
| 44 |
+
else:
|
| 45 |
+
sentence_scores[sent] += word_frequencies[word]
|
| 46 |
+
summary_sentences = heapq.nlargest(7, sentence_scores, key=sentence_scores.get)
|
| 47 |
+
summary = ' '.join(summary_sentences)
|
| 48 |
+
return summary
|
| 49 |
+
|
| 50 |
+
# Streamlit UI
|
| 51 |
+
st.title("Text Summarization App")
|
| 52 |
+
|
| 53 |
+
# User input for URL
|
| 54 |
+
url = st.text_input("Enter URL:")
|
| 55 |
+
|
| 56 |
+
if url:
|
| 57 |
+
try:
|
| 58 |
+
summary = summarize_text(url)
|
| 59 |
+
st.subheader("Summary")
|
| 60 |
+
st.write(summary)
|
| 61 |
+
except Exception as e:
|
| 62 |
+
st.error(f"An error occurred: {e}")
|