app.py

import pandas as pd
import numpy as np
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.neighbors import KNeighborsClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
import gradio as gr

# Load and preprocess the data
file_path = 'homework01_text_data_group16.csv'  # Update with your file path
data = pd.read_csv(file_path)
sentences = data['reviews']
labels = data['class']

# Split the data
sentence_train, sentence_test, y_train, y_test = train_test_split(
    sentences, labels, test_size=0.2, random_state=42
)

# Bag-of-Words vectorization
vectorizer = CountVectorizer()
X_train = vectorizer.fit_transform(sentence_train)
X_test = vectorizer.transform(sentence_test)

# Train models
knn_model = KNeighborsClassifier(n_neighbors=3).fit(X_train, y_train)
logistic_model = LogisticRegression().fit(X_train, y_train)
svm_model = SVC(kernel='linear', probability=True).fit(X_train, y_train)
rf_model = RandomForestClassifier(n_estimators=100).fit(X_train, y_train)

# Gradio app function
def predict_sentiment(text, model_name):
    vectorized_text = vectorizer.transform([text])
    if model_name == 'KNN':
        probabilities = knn_model.predict_proba(vectorized_text)[0]
    elif model_name == 'Logistic Regression':
        probabilities = logistic_model.predict_proba(vectorized_text)[0]
    elif model_name == 'SVM':
        probabilities = svm_model.predict_proba(vectorized_text)[0]
    elif model_name == 'Random Forest':
        probabilities = rf_model.predict_proba(vectorized_text)[0]
    else:
        return "Invalid Model Selection"
    
    prediction = "Positive" if probabilities[1] > probabilities[0] else "Negative"
    prob_positive = round(probabilities[1] * 100, 2)
    prob_negative = round(probabilities[0] * 100, 2)
    
    return f"Prediction: {prediction}", f"Positive: {prob_positive}%\nNegative: {prob_negative}%"

# Define the Gradio interface
interface = gr.Interface(
    fn=predict_sentiment,
    inputs=["text", gr.Dropdown(["KNN", "Logistic Regression", "SVM", "Random Forest"])],
    outputs=[gr.Textbox(label="Prediction"), gr.Textbox(label="Details (Probabilities)")],
    title="Sentiment Analysis with Multiple Models",
    description="Enter a text review and select a model to predict sentiment and view probabilities."
)

# Launch the app
if __name__ == "__main__":
    interface.launch()