File size: 3,599 Bytes
8a890e9 c13fd5d d9f1d02 8a890e9 c13fd5d aab69e8 c13fd5d f0c7269 c13fd5d f0c7269 946ddbf aab69e8 c13fd5d 8a890e9 c13fd5d 8a890e9 c13fd5d d9f1d02 aab69e8 8a890e9 c13fd5d 8a890e9 d9f1d02 4a3dc04 aab69e8 4a3dc04 8a890e9 aab69e8 |
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 90 91 92 93 94 95 96 97 98 99 100 101 102 103 |
import streamlit as st
import nltk
from nltk.corpus import stopwords
from nltk.cluster.util import cosine_distance
import numpy as np
import networkx as nx
from googletrans import Translator
# Download NLTK resources
nltk.download('punkt')
nltk.download('stopwords')
# Function to read and preprocess the article
def read_article(article):
sentences = nltk.sent_tokenize(article)
sentences = [sentence for sentence in sentences if len(sentence) > 10] # Filter out very short sentences
return sentences
# Function to compute sentence similarity based on cosine similarity
def sentence_similarity(sent1, sent2, stopwords):
words1 = nltk.word_tokenize(sent1)
words2 = nltk.word_tokenize (sent2)
words1 = [word.lower() for word in words1 if word.isalnum()]
words2 = [word.lower() for word in words2 if word.isalnum()]
all_words = list(set(words1 + words2))
vector1 = [0] * len(all_words)
vector2 = [0] * len(all_words)
for word in words1:
if word in stopwords:
continue
vector1[all_words.index(word)] += 1
for word in words2:
if word in stopwords:
continue
vector2[all_words.index(word)] += 1
return 1 - cosine_distance(vector1, vector2)
# Function to create a similarity matrix of sentences
def build_similarity_matrix(sentences, stopwords):
similarity_matrix = np.zeros((len(sentences), len(sentences)))
for i in range(len(sentences)):
for j in range(len(sentences)):
if i == j: # Skip comparing a sentence to itself
continue
similarity_matrix[i][j] = sentence_similarity(sentences[i], sentences[j], stopwords)
return similarity_matrix
# Function to generate the article summary
def generate_summary(article, top_n=5):
sentences = read_article(article)
stop_words = set(stopwords.words('english'))
sentence_similarity_matrix = build_similarity_matrix(sentences, stop_words)
# Create a graph from the similarity matrix
sentence_similarity_graph = nx.from_numpy_array(sentence_similarity_matrix)
# Use the PageRank algorithm to rank the sentences
scores = nx.pagerank(sentence_similarity_graph)
# Sort the sentences by score
ranked_sentences = sorted(((scores[i], sentence) for i, sentence in enumerate(sentences)), reverse=True)
# Get the top N sentences as the summary
summary = " ".join([sentence for _, sentence in ranked_sentences[:top_n]])
return summary
# Streamlit web app
st.title("Article Summarizer")
user_article = st.text_area("Enter your article here:")
translate = st.checkbox("Translate Summary")
target_language = st.selectbox("Select Target Language", ["English", "French", "Spanish", "German"])
if st.button("Summarize"):
if user_article:
summary = generate_summary(user_article)
st.subheader("Summary:")
st.write(summary)
if translate:
if target_language == "English":
target_language_code = "en"
elif target_language == "French":
target_language_code = "fr"
elif target_language == "Spanish":
target_language_code = "es"
elif target_language == "German":
target_language_code = "de"
translator = Translator()
translated_summary = translator.translate(summary, dest=target_language_code)
st.subheader("Translated Summary:")
st.write(translated_summary.text)
else:
st.warning("Please enter an article to summarize.") |