Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -28,9 +28,9 @@ def jieba_tokenizer(text):
|
|
| 28 |
vectorizer = CountVectorizer(tokenizer=jieba_tokenizer)
|
| 29 |
kw_model = KeyBERT()
|
| 30 |
|
| 31 |
-
# Extract keywords
|
| 32 |
-
def extract_keywords(doc):
|
| 33 |
-
keywords = kw_model.extract_keywords(doc, vectorizer=vectorizer)
|
| 34 |
return keywords
|
| 35 |
|
| 36 |
# Plot keywords
|
|
@@ -48,7 +48,7 @@ def plot_keywords(keywords, title):
|
|
| 48 |
return '/tmp/keywords_plot.png'
|
| 49 |
|
| 50 |
# Function to scrape content and extract keywords
|
| 51 |
-
def scrape_and_extract(url):
|
| 52 |
response = requests.get(url)
|
| 53 |
response.encoding = 'utf-8'
|
| 54 |
soup = BeautifulSoup(response.text, 'html.parser')
|
|
@@ -56,7 +56,7 @@ def scrape_and_extract(url):
|
|
| 56 |
content_div = soup.find('div', {'class': 'caas-body'})
|
| 57 |
paragraphs = content_div.find_all('p')
|
| 58 |
content = '\n'.join([p.text.strip() for p in paragraphs])
|
| 59 |
-
keywords = extract_keywords(content)
|
| 60 |
plot_path = plot_keywords(keywords, "Keyword Extraction Results")
|
| 61 |
return title, content, keywords, plot_path
|
| 62 |
|
|
@@ -67,10 +67,11 @@ st.title("π Professional Keyword Extraction Tool")
|
|
| 67 |
st.write("Extracts keywords from a given URL and displays a bar chart of the keywords with their respective scores.")
|
| 68 |
|
| 69 |
url = st.text_input("π Enter the article URL here:")
|
|
|
|
| 70 |
|
| 71 |
if st.button("Extract Keywords"):
|
| 72 |
if url:
|
| 73 |
-
title, content, keywords, plot_path = scrape_and_extract(url)
|
| 74 |
|
| 75 |
st.subheader("π Article Title")
|
| 76 |
st.write(title)
|
|
@@ -86,5 +87,3 @@ if st.button("Extract Keywords"):
|
|
| 86 |
st.image(plot_path)
|
| 87 |
else:
|
| 88 |
st.warning("Please enter a URL to extract keywords.")
|
| 89 |
-
|
| 90 |
-
|
|
|
|
| 28 |
vectorizer = CountVectorizer(tokenizer=jieba_tokenizer)
|
| 29 |
kw_model = KeyBERT()
|
| 30 |
|
| 31 |
+
# Extract keywords using MMR
|
| 32 |
+
def extract_keywords(doc, diversity):
|
| 33 |
+
keywords = kw_model.extract_keywords(doc, vectorizer=vectorizer, use_mmr=True, diversity=diversity)
|
| 34 |
return keywords
|
| 35 |
|
| 36 |
# Plot keywords
|
|
|
|
| 48 |
return '/tmp/keywords_plot.png'
|
| 49 |
|
| 50 |
# Function to scrape content and extract keywords
|
| 51 |
+
def scrape_and_extract(url, diversity):
|
| 52 |
response = requests.get(url)
|
| 53 |
response.encoding = 'utf-8'
|
| 54 |
soup = BeautifulSoup(response.text, 'html.parser')
|
|
|
|
| 56 |
content_div = soup.find('div', {'class': 'caas-body'})
|
| 57 |
paragraphs = content_div.find_all('p')
|
| 58 |
content = '\n'.join([p.text.strip() for p in paragraphs])
|
| 59 |
+
keywords = extract_keywords(content, diversity)
|
| 60 |
plot_path = plot_keywords(keywords, "Keyword Extraction Results")
|
| 61 |
return title, content, keywords, plot_path
|
| 62 |
|
|
|
|
| 67 |
st.write("Extracts keywords from a given URL and displays a bar chart of the keywords with their respective scores.")
|
| 68 |
|
| 69 |
url = st.text_input("π Enter the article URL here:")
|
| 70 |
+
diversity = st.slider("Adjust Diversity (0.0: Most Relevant, 1.0: Most Diverse)", 0.0, 1.0, 0.5, step=0.01)
|
| 71 |
|
| 72 |
if st.button("Extract Keywords"):
|
| 73 |
if url:
|
| 74 |
+
title, content, keywords, plot_path = scrape_and_extract(url, diversity)
|
| 75 |
|
| 76 |
st.subheader("π Article Title")
|
| 77 |
st.write(title)
|
|
|
|
| 87 |
st.image(plot_path)
|
| 88 |
else:
|
| 89 |
st.warning("Please enter a URL to extract keywords.")
|
|
|
|
|
|