Upload 8 files

app.py

@@ -0,0 +1,22 @@
+import streamlit as st
+import pandas as pd
+from eda import display_eda
+from prediction import predict_and_strategy
+
+# Load the data
+data = pd.read_csv('threads_reviews.csv')
+
+st.title("Sentiment Analysis and Business Strategy")
+
+# EDA Section
+st.header("Exploratory Data Analysis")
+if st.checkbox("Show EDA", False):  # Checkbox to toggle EDA display
+    display_eda(data)
+
+# Prediction Section
+st.header("Prediction")
+user_input = st.text_area("Enter text for sentiment analysis:", "")
+if st.button("Analyze"):
+    sentiment, strategy = predict_and_strategy(user_input)
+    st.write(f"Sentiment: {sentiment}")
+    st.write(f"Strategy: {strategy}")

eda.py

@@ -0,0 +1,23 @@
+import streamlit as st
+import pandas as pd
+import matplotlib.pyplot as plt
+from wordcloud import WordCloud
+
+def display_eda(data):
+    # Distribution of sentiments
+    st.subheader("Distribution of Sentiments")
+    sentiment_counts = data['sentiment'].value_counts()
+    st.bar_chart(sentiment_counts)
+    
+    # Word cloud for each sentiment
+    st.subheader("Word Clouds for Sentiments")
+    sentiments = data['sentiment'].unique()
+    for sentiment in sentiments:
+        st.write(f"Word Cloud for {sentiment}")
+        subset = data[data['sentiment'] == sentiment]
+        text = " ".join(review for review in subset['processed_review'])
+        wordcloud = WordCloud(max_words=100, background_color="white").generate(text)
+        plt.figure()
+        plt.imshow(wordcloud, interpolation="bilinear")
+        plt.axis("off")
+        st.pyplot()

my_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e57bd45f6afda98272478433c8df6e4ed7a2d19b41c9f8b4c8de54ff0da7264
+size 3911040

prediction.py

@@ -0,0 +1,17 @@
+from transformers import pipeline
+
+nlp = pipeline("sentiment-analysis")
+
+def predict_and_strategy(text):
+    result = nlp(text)
+    sentiment = result[0]['label']
+
+    # Provide strategy based on sentiment
+    if sentiment == "POSITIVE":
+        strategy = "Engage with these customers to make them brand ambassadors."
+    elif sentiment == "NEUTRAL":
+        strategy = "Try to find out what's missing and engage more with these customers."
+    else:
+        strategy = "Address the concerns of these customers immediately."
+    
+    return sentiment, strategy

preprocessing_data.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f2a2425763e25b1ad53362b6c2fe6b9833611484f11058b70d609ef0c18e07e
+size 90

requirements.txt

@@ -0,0 +1,23 @@
+# Basic libraries
+numpy
+pandas
+matplotlib
+seaborn
+
+# Machine Learning and Deep Learning
+tensorflow
+keras
+scikit-learn
+
+# Natural Language Processing
+transformers  # from Hugging Face
+tokenizers    # often used alongside transformers
+
+# Web App
+streamlit
+
+# Miscellaneous
+requests
+
+# Depending on your specific needs or the deployment platform, you might also need:
+gunicorn     # WSGI HTTP Server for UNIX, often used for deploying Flask and Streamlit apps on platforms like Heroku

tokenizer.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ea16cd2ff6a9a38106463fb4e07bafbd57ed8180cac5d7cd07a4153eb1effd7
+size 574685