app.py

import streamlit as st
import pandas as pd
import joblib
from gensim import corpora, models
from PIL import Image

# Load the saved models and data
dictionary = joblib.load('doc2bow.sav')  # Load the dictionary
lda_model = joblib.load('ldamodel.sav')  # Load the LDA model

# Function to preprocess input text and get topic distribution
def preprocess(text):
    # Define your preprocessing logic here, as it was in your original code
    result = []
    for token in gensim.utils.simple_preprocess(text):
        if token not in gensim.parsing.preprocessing.STOPWORDS and token not in newStopWords and len(token) > 3:
            result.append(lemmatize_stemming(token))
    return result

def get_topics(text):
    bow_vector = dictionary.doc2bow(preprocess(text))
    topics = lda_model[bow_vector]
    return topics

# Function to get top keywords for a topic
def get_top_keywords(topic, num_keywords=10):
    topic = lda_model.show_topic(topic, topn=num_keywords)
    keywords = [f"{word} ({weight:.3f})" for word, weight in topic]
    return keywords

# Streamlit app
def main():
    st.title("Web Berita Topic Clustering Untuk Program Kedaireka UMKM📰")

     # Sidebar with title and description
    st.sidebar.title("Topic Clustering")
    st.sidebar.write("Discover topics in news articles.")
    
    # Input text area for user to enter their text
    user_input = st.text_area("Enter your text here:", "")
    
    # Submit button
    if st.button("Submit"):
        if user_input:
            zkata =[]
            # Masukkan user input ke dalam zkata
            zkata.append(user_input)
    
            # Gabungkan zkata menjadi satu string
            string = ' '.join([str(item) for item in zkata])
            tampung_kata = string
    
            # Lakukan pemrosesan pada `tampung_kata` dan dictionary
            bow_vector = dictionary.doc2bow(preprocess(tampung_kata))
    
            # Analisis topik
            st.subheader("🔥Top Topics🔥")
            for index, score in sorted(lda_model[bow_vector], key=lambda tup: -1 * tup[1]):
                st.write(f"Score: {score}\t Topic: {index + 1} | {lda_model.print_topic(index, 10)}")

    # Add a footer
    st.sidebar.markdown("---")
    st.sidebar.write("© 2023 Web Berita Topic Clustering")

if __name__ == "__main__":
    main()