lib/sentiment_analyser.py

import os
import string
from collections import Counter
from datetime import datetime, timezone, timedelta
from pathlib import Path

import nltk
from nltk.corpus import stopwords
from nltk.sentiment.vader import SentimentIntensityAnalyzer
from nltk.stem import WordNetLemmatizer
from nltk.tokenize import word_tokenize

from utils import *

datetime_format = "%d/%m/%Y %H:%M:%S"
ist_offset = timedelta(hours=5, minutes=30)
def now():
    utc_time = datetime.now(timezone.utc)
    ist_time = utc_time.astimezone(timezone(ist_offset))
    return ist_time.strftime(datetime_format)

class SentimentAnalyser:
    def __init__(self):
        nltk.download('punkt')
        nltk.download('stopwords')
        nltk.download('wordnet')
        self.emotions = Path("utils/emotions.txt").resolve()
    
    def sentiment(self, text):
        prompt = text
        lower_case = text.lower()
        cleaned_text = lower_case.translate(str.maketrans('', '', string.punctuation))

        # Using word_tokenize because it's faster than split()
        tokenized_words = word_tokenize(cleaned_text, "english")

        # Removing Stop Words
        final_words = []
        for word in tokenized_words:
            if word not in stopwords.words("english"):
                final_words.append(word)
        
        # Lemmatization - From plural to single + base form of a word (example better -> good)
        lemma_words = []
        for word in final_words:
            word = WordNetLemmatizer().lemmatize(word)
            lemma_words.append(word)
        
        emotion_list = []
        with open(self.emotions) as f:
            for line in f:
                clear_line = line.replace("\n", "").replace(",", "").replace("'", "").replace(" ", "").strip()
                word, emotion = clear_line.split(":")

                if word in lemma_words:
                    emotion_list.append(emotion)
        
        print(f"[{now()}] Emotion List:", emotion_list)
        if not len(emotion_list):
            print(f"[{now()}] No emotion could be extracted.")
            return None
        
        emotions_count = Counter(emotion_list)
        print(f"[{now()}] Emotions Count:", emotions_count)

        common = emotions_count.most_common(1)
        print(f"[{now()}] Common Emotions:", common)

        sentiment, val = common[0]
        print(f"[{now()}] Emotion:", sentiment)
        return sentiment