Negation in Sentimental Analysis

I'm trying use text normalization, and prepend the prefix NOT_ to every word after a token of logical negation, specifically for the sentence "didn't like this movie , but I."

By doing this, it creates the output:

"didn't NOT_like NOT_this NOT_movie , but I"

however I'm getting the output:

"didnt like NOT_didnt NOT_like this NOT_like NOT_this NOT_didnt NOT_like NOT_this movie NOT_this NOT_movie NOT_like NOT_this NOT_movie but NOT_movie NOT_but NOT_this NOT_movie NOT_but i NOT_but NOT_i NOT_movie NOT_but NOT_i".

I feel like I must be doing something obviously wrong--I'll past my code in below. Not being able to figure out what I'm doing wrong is driving me crazy.

def preprocess_text(text):    text = text.lower()    text = re.sub(r'[^\w\s\']', '', text)    return text    text = text.lower()    text = re.sub(r'[^\w\s\']', '', text)    return texttext = "didn’t like this movie, but I"def negify(text):    negation_terms = {"not", "n't", "no"}    punctuation_marks = {".", ",", "!", "?", ";"}    words = text.split()    negated_text = []    negation_active = False    result = []    for word in words:        word_stripped = word.strip('.,?!:;').lower()         if negation_active:            negated_text.append("not_" + word_stripped)        else:            negated_text.append(word_stripped)        result.append(word_stripped)        if len(negated_text) > 1:            bigram = " ".join(negated_text[-2:])            result.append(bigram)            if len(negated_text) > 2:                trigram = " ".join(negated_text[-3:])                result.append(trigram)        if any(neg in word.lower() for neg in negation_terms):            negation_active = not negation_active        if any(p in word for p in punctuation_marks):            negation_active = False    return " ".join(result)def binarize(text):    words = text.split()    binarized_words = []    negation_active = False    for word in words:        if word.lower() == "didn't":            binarized_words.append(word)            words = text.split()    unique_words = set()    binarized_words = []    for word in words:        if word.lower() not in unique_words:            unique_words.add(word.lower())            binarized_words.append(word)        else:            binarized_words.append('NOT_'+ word)    binarized_text = ''.join(binarized_words)    return binarized_texttext = preprocess_text(text)text = negify(text)binarized_output = binarize(text)print(binarized_output)``didnt like NOT_didnt NOT_like this NOT_like NOT_this NOT_didnt NOT_like NOT_this movie NOT_this NOT_movie NOT_like NOT_this NOT_movie but NOT_movie NOT_but NOT_this NOT_movie NOT_but i NOT_but NOT_i NOT_movie NOT_but NOT_i