SentimentAnalyzer.py

import tkinter as tk
import nltk
import requests
import re
from word2number import w2n
from tkinter import ttk
from PIL import Image, ImageTk
from nltk.sentiment import SentimentIntensityAnalyzer
from nltk.tokenize import word_tokenize

nltk.download('averaged_perceptron_tagger')
nltk.download('vader_lexicon')

analyzer = SentimentIntensityAnalyzer()


def convert_numbers(sentence):
    words = sentence.split()
    for i in range(len(words)):
        try:
            words[i] = str(w2n.word_to_num(words[i]))
        except ValueError:
            pass
    return ' '.join(words)


def get_word_role(word, sentence):
    words = word_tokenize(sentence)
    tagged_words = nltk.pos_tag(words)

    for i, tagged_word in enumerate(tagged_words):
        if tagged_word[0].lower() == word.lower():
            if word.lower() == "am" and ((i > 0 and re.search(r'\d', tagged_words[i - 1][0])) or
                                         (i < len(tagged_words) - 1 and re.search(r'\d', tagged_words[i + 1][0]))):
                return "Time"
            else:
                return tagged_word[1]

    return 'Role Not found'


def analyze_sentiment():
    text = input_field.get("1.0", "end").strip()
    text = convert_numbers(text)
    if text and text != "Please enter some text...":
        words = text.split()
        if len(words) <= 20:
            sentiment_score = analyzer.polarity_scores(text)
            compound_score = sentiment_score['compound']
            if compound_score >= 0.05:
                sentiment = "This is a Positive sentiment."
            elif compound_score <= -0.05:
                sentiment = "This is a Negative sentiment."
            else:
                sentiment = "This is a Neutral sentiment."

            explanations = []
            roles = []
            for i, word in enumerate(words):
                role = get_word_role(word, text)
                roles.append(role)
                if word.lower() == 'am' and i > 0 and words[i - 1].isdigit():
                    explanation = "Referring to time"
                elif word.lower() == 'am' and 'VB' in role:
                    explanation = "First person singular present of 'be'"
                else:
                    explanation = get_word_explanation(word)
                explanations.append(explanation)

            output_text = sentiment
            output_text += ' It is because: \n'
            for word, explanation, role in zip(words, explanations, roles):
                output_text += f'{word} ({role}): {explanation}\n'

            output_label.config(text=output_text)
        else:
            output_label.config(text="Please enter no more than 20 words.")
    else:
        output_label.config(text="Please enter some text.")



def create_gradient(width, height):
    base = Image.new('RGB', (width, height), '#2C3E50')
    top = Image.new('RGB', (width, height), '#4CA1AF')
    mask = Image.new('L', (width, height))
    mask_data = []
    for y in range(height):
        mask_data.extend(list(int(255 * (x / width)) for x in range(width)))
    mask.putdata(mask_data)
    base.paste(top, (0, 0), mask)
    return base


gradient = create_gradient(900, 700)


def get_word_explanation(word):
    url = f'https://api.dictionaryapi.dev/api/v2/entries/en/{word}'
    response = requests.get(url)
    if response.status_code == 200:
        data = response.json()
        if len(data) > 0 and 'meanings' in data[0]:
            meanings = data[0]['meanings']
            if len(meanings) > 0 and 'definitions' in meanings[0]:
                definitions = meanings[0]['definitions']
                if len(definitions) > 0 and 'definition' in definitions[0]:
                    return definitions[0]['definition']
    return 'Explanation Not found'


def reset_output_label():
    output_label.config(background='white', font=('Arial', 11))


window = tk.Tk()
window.title("Sentiment Analyzer")
window.geometry("900x700")

background_image = ImageTk.PhotoImage(gradient)
background_label = tk.Label(window, image=background_image)
background_label.place(x=0, y=0, relwidth=1, relheight=1)

title_label_font = ('Lora', 24, 'bold')
title_label = tk.Label(window, text="Sentiment Analyzer", font=title_label_font, fg='black')
title_label.pack(pady=10)

input_field_font = ('Consolas', 14)
input_field = tk.Text(window, height=10, width=70, font=input_field_font, fg='gray')
input_field.insert("end", "Please enter some text...")


def add_prompt(event):
    if input_field.get("1.0", "end").strip() == "Please enter some text...":
        input_field.delete("1.0", "end")
        input_field.config(fg="black")


input_field.bind("<Button-1>", add_prompt)
input_field.pack(pady=20)

style = ttk.Style()
style.configure('my.TButton', font=('Arial', 16), foreground='black')
analyze_button = ttk.Button(window, text="Analyze", command=analyze_sentiment, style='my.TButton')
analyze_button.pack()

output_label_font = ('Roboto', 14)
output_label = tk.Label(window, text="", wraplength=750, justify='left', font=output_label_font, bg='white', fg='black')
output_label.pack(pady=20, padx=20)

footer_label_font = ('Montserrat', 10)
footer_label = tk.Label(window, text="Copyright © Khandakar Sayeem. All Rights Reserved. 2023", font=footer_label_font)
footer_label.pack(side=tk.BOTTOM, pady=10)

window.mainloop()