comment_analyzer.py

import gradio as gr
import pandas as pd
from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer
from sklearn.model_selection import train_test_split
from sklearn.svm import SVR


data = pd.read_csv("modeled_data.csv")
analyzer = SentimentIntensityAnalyzer()


def sample_model(df, regressor):
    X = df.drop("rate",axis=1)
    y = df["rate"]
    
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state=1)
    
    model = regressor
    model.fit(X_train, y_train)
    
    return model


def calculate_sentiments(comment, model):
    
    negative_score = analyzer.polarity_scores(comment)["neg"]
    neutral_score = analyzer.polarity_scores(comment)["neu"]
    positive_score = analyzer.polarity_scores(comment)["pos"]
    compound_score = analyzer.polarity_scores(comment)["compound"]
    rate_pred = model.predict([[negative_score, neutral_score, positive_score, compound_score]])
    
    return round(negative_score,2), round(neutral_score,2), round(positive_score,2), round(compound_score,2), round(rate_pred[0],2)

    
def take_input(comment):
    
    cons_tuned_svr = sample_model(data, SVR(C=3, kernel="rbf", tol=0.001))
    return calculate_sentiments(comment, cons_tuned_svr)


with gr.Blocks() as demo:
    gr.Markdown("# AIN311 Project P05 - MOOC Recommendation")
    gr.Markdown("## Generating a Rating from User Comment")
    with gr.Column():
        gr.Markdown("""
                    ##### Thanks for your interest and taking your time.
                    ##### Tell us about your personal experience enrolling in this course. Was it the right match for you?
                    """)
        input_comment = gr.Textbox(placeholder="Write your comment here...", show_label = False, lines=2)
        button = gr.Button("What is the Rating I Have Given? Click me to Learn", variant="secondary").style(full_width=True)
        with gr.Row():
            with gr.Column():
                gr.Markdown("#### Generated Rating from Your Comment")
                rating = gr.Number().style(show_label=False)
            with gr.Column():
                gr.Markdown("#### Sentiment Scores of Your Comment")
                with gr.Row():
                    negscore = gr.Number(label="Negativity Score")
                    neuscore = gr.Number(label="Neutrality Score")
                    posscore = gr.Number(label="Positivity Score")
                    compscore = gr.Number(label="Compound Score")

        gr.Markdown("Example comments has taken from https://www.udemy.com/course/statistics-for-data-science-and-business-analysis/")
        gr.Examples(
                     [["Not really a basic course. Goes too fast and sometimes the explanations aren't clear. The solutions to the exercises are not explained, and should be. Not much better than the other course I started and didn't finish.(3)"], 
                      ["Presentation style was bad. Too much detail on the simpler topics, and glossed over some of the more complicated ones.(2)"], 
                      ["This course is the worst Course i have ever watched (1)"],
                      ["The course is really great! The didatic in explain all concepts and pratical examples are amazing. Better than brazilian universities.(4)"],
                      ["The best!!!!!!!!!!(5)"],
                      ["Excelent description and view. The exercises were prepared very carefully. I suggest it.(5)"],
                      ["Yes, This course is a very good way to update/ refresh statistical knowledge (2)"],
                      ["explaination is good but practical examples are not so good (2.5)"],
                      ["Thanks for content. Good to know and understand the things easily. (3.5)"],
                      ["good (3)"],
                      ["The course navigation is very bad ..It is very tedious to navigate the course.(5)"]],
                     [input_comment],
                     [[negscore, neuscore, posscore, compscore, rating]],
                     fn=take_input
        )
    button.click(fn=take_input, inputs=input_comment, outputs=[negscore, neuscore, posscore, compscore, rating])

demo.launch()