Commit
·
3fec030
1
Parent(s):
309b488
Update app.py
Browse files
app.py
CHANGED
@@ -3,6 +3,15 @@ import streamlit as st
|
|
3 |
from google.oauth2 import service_account
|
4 |
from google.cloud import language_v1
|
5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
# Sidebar content
|
7 |
st.sidebar.title("About This Tool")
|
8 |
st.sidebar.markdown("This tool leverages Google's NLP technology for entity analysis.")
|
@@ -12,17 +21,14 @@ st.sidebar.markdown("""
|
|
12 |
2. **User Input**: Enter the text you want to analyze.
|
13 |
3. **Analyze**: Click the 'Analyze' button.
|
14 |
4. **View Results**: See the identified entities and their details.
|
15 |
-
|
16 |
-
To do:
|
17 |
-
https://www.linkedin.com/pulse/seo-content-writing-how-optimize-entity-salience-emmanuel-dan-awoh/
|
18 |
""")
|
19 |
|
20 |
# Header and intro
|
21 |
st.title("Google Cloud NLP Entity Analyzer")
|
22 |
st.write("This tool analyzes text to identify entities such as people, locations, organizations, and events")
|
23 |
st.write("Entity salience scores are always relative to the analysed text. In natural language processing, a salience score is always a prediction of what a human would consider to be the most important entities in the same text. A number of textual features contribute to the salience score.")
|
|
|
24 |
def sample_analyze_entities(text_content):
|
25 |
-
# Assuming service_account_info is set in your Streamlit secrets
|
26 |
service_account_info = json.loads(st.secrets["google_nlp"])
|
27 |
credentials = service_account.Credentials.from_service_account_info(
|
28 |
service_account_info, scopes=["https://www.googleapis.com/auth/cloud-platform"]
|
@@ -33,8 +39,11 @@ def sample_analyze_entities(text_content):
|
|
33 |
encoding_type = language_v1.EncodingType.UTF8
|
34 |
|
35 |
response = client.analyze_entities(request={"document": document, "encoding_type": encoding_type})
|
|
|
|
|
|
|
36 |
|
37 |
-
st.write(f"
|
38 |
st.write("---")
|
39 |
|
40 |
for i, entity in enumerate(response.entities):
|
|
|
3 |
from google.oauth2 import service_account
|
4 |
from google.cloud import language_v1
|
5 |
|
6 |
+
# Function to count entities with 'mid' and '/g/' in their metadata
|
7 |
+
def count_entities(entities):
|
8 |
+
count = 0
|
9 |
+
for entity in entities:
|
10 |
+
metadata = entity.metadata
|
11 |
+
if 'mid' in metadata and '/g/' in metadata['mid']:
|
12 |
+
count += 1
|
13 |
+
return count
|
14 |
+
|
15 |
# Sidebar content
|
16 |
st.sidebar.title("About This Tool")
|
17 |
st.sidebar.markdown("This tool leverages Google's NLP technology for entity analysis.")
|
|
|
21 |
2. **User Input**: Enter the text you want to analyze.
|
22 |
3. **Analyze**: Click the 'Analyze' button.
|
23 |
4. **View Results**: See the identified entities and their details.
|
|
|
|
|
|
|
24 |
""")
|
25 |
|
26 |
# Header and intro
|
27 |
st.title("Google Cloud NLP Entity Analyzer")
|
28 |
st.write("This tool analyzes text to identify entities such as people, locations, organizations, and events")
|
29 |
st.write("Entity salience scores are always relative to the analysed text. In natural language processing, a salience score is always a prediction of what a human would consider to be the most important entities in the same text. A number of textual features contribute to the salience score.")
|
30 |
+
|
31 |
def sample_analyze_entities(text_content):
|
|
|
32 |
service_account_info = json.loads(st.secrets["google_nlp"])
|
33 |
credentials = service_account.Credentials.from_service_account_info(
|
34 |
service_account_info, scopes=["https://www.googleapis.com/auth/cloud-platform"]
|
|
|
39 |
encoding_type = language_v1.EncodingType.UTF8
|
40 |
|
41 |
response = client.analyze_entities(request={"document": document, "encoding_type": encoding_type})
|
42 |
+
|
43 |
+
# Count the entities with 'mid' and '/g/' in their metadata
|
44 |
+
entity_count = count_entities(response.entities)
|
45 |
|
46 |
+
st.write(f"We found {len(response.entities)} entities - {entity_count} meet your criteria")
|
47 |
st.write("---")
|
48 |
|
49 |
for i, entity in enumerate(response.entities):
|