Text_analysis_and_metadata_app / Text_analysis.py
UjjwalVIT's picture
strings
56d18f3
raw
history blame
6.7 kB
import streamlit as st
import pandas as pd
import streamlit.components.v1 as stc
import docx2txt
# NLP Package-used for text analysis
import nltk
from nltk.tokenize import word_tokenize
from nltk.tag import pos_tag
from nltk.stem import WordNetLemmatizer
from nltk.corpus import stopwords
# from nltk import ne_chunk
from nltk.tag import StanfordNERTagger
from collections import Counter
from textblob import TextBlob
import seaborn as sns
import matplotlib.pyplot as plt
from wordcloud import WordCloud
import base64
import time
from app_utils import *
HTML_BANNER = """
<div style="background-color:green;padding:10px;border-radius:10px">
<h1 style="color:white;text-align:center;">Text Analysis App </h1>
</div>
"""
def text_analysis():
stc.html(HTML_BANNER)
menu=['Text-analysis','Upload_Files']
choice=st.sidebar.selectbox('Menu',menu)
if choice=='Text-analysis':
st.subheader('Analyse Text')
text=st.text_area("Enter the text to anlayze")
if (st.button("Analyze")):
st.success("Success")
with st.expander('Original Text'):
st.write(text)
with st.expander('Text Analysis'):
token_analysis=nlp_analysis(text)
st.dataframe(token_analysis)
with st.expander('Entitites'):
entity_result=find_entities(text)
stc.html(entity_result, height=100, scrolling=True)
col1,col2=st.columns(2)
with col1:
with st.expander("Word Stats"):
st.info("Word Statistics")
docx = nt.TextFrame(text)
st.write(docx.word_stats())
with st.expander("Top keywords"):
keywords=get_most_common_tokens(text)
st.write(keywords)
with st.expander('Tagged Keywords'):
data= pos_tag(word_tokenize(text))
st.dataframe(data)
visualize_tags=tag_visualize(data)
stc.html(visualize_tags,scrolling=True)
with st.expander("Sentiment"):
sent_result=get_semantics(text)
st.write(sent_result)
with col2:
with st.expander("Plot word freq"):
try:
fig, ax = plt.subplots()
most_common_tokens = dict(token_analysis["Token"].value_counts())
sns.countplot(data=token_analysis[token_analysis["Token"].isin(most_common_tokens)], x="Token", ax=ax)
ax.set_xlabel('PoS')
ax.set_ylabel('Frequency')
ax.tick_params(axis='x' , rotation=45)
st.pyplot(fig)
except:
st.warning('Insufficient data')
with st.expander("Plot part of speech"):
try:
fig, ax = plt.subplots()
most_common_tokens = dict(token_analysis["Position"].value_counts())
sns.countplot(data=token_analysis[token_analysis["Position"].isin(most_common_tokens)], x="Position", ax=ax)
ax.set_xlabel('PoS')
ax.set_ylabel('Frequency')
ax.tick_params(axis='x' , rotation=45)
st.pyplot(fig)
except:
st.warning('Insufficient data')
with st.expander("Plot word cloud"):
try:
plot_wordcloud(text)
except:
st.warning('Insufficient data')
with st.expander('Download Results'):
file_download(token_analysis)
elif choice == 'Upload_Files':
text_file = st.file_uploader('Upload Files', type=['docx'])
if text_file is not None:
if text_file.type == 'text/plain':
text = str(text_file.read(), "utf-8")
else:
text = docx2txt.process(text_file)
if (st.button("Analyze")):
with st.expander('Original Text'):
st.write(text)
with st.expander('Text Analysis'):
token_analysis = nlp_analysis(text)
st.dataframe(token_analysis)
with st.expander('Entities'):
entity_result = find_entities(text)
stc.html(entity_result, height=100, scrolling=True)
col1, col2 = st.columns(2)
with col1:
with st.expander("Word Stats"):
st.info("Word Statistics")
docx = nt.TextFrame(text)
st.write(docx.word_stats())
with st.expander("Top keywords"):
keywords = get_most_common_tokens(text)
st.write(keywords)
with st.expander("Sentiment"):
sent_result = get_semantics(text)
st.write(sent_result)
with col2:
with st.expander("Plot word freq"):
fig, ax = plt.subplots()
num_tokens = 10 # Adjust the number of tokens to display as desired
most_common_tokens = dict(token_analysis["Token"].value_counts().head(num_tokens))
sns.countplot(data=token_analysis[token_analysis["Token"].isin(most_common_tokens)], x="Token", ax=ax)
ax.set_xlabel('Token')
ax.set_ylabel('Frequency')
ax.tick_params(axis='x', rotation=45)
st.pyplot(fig)
with st.expander("Plot part of speech"):
fig, ax = plt.subplots()
most_common_tokens = dict(token_analysis["Position"].value_counts())
sns.countplot(data=token_analysis[token_analysis["Position"].isin(most_common_tokens)], x="Position", ax=ax)
ax.set_xlabel('PoS')
ax.set_ylabel('Frequency')
ax.tick_params(axis='x', rotation=45)
st.pyplot(fig)
with st.expander("Plot word cloud"):
plot_wordcloud(text)
with st.expander('Download Results'):
file_download(token_analysis)