def caclulate_jaccard_similarity(conclusion, premise):
"""
Calculate jacard similarity between conclusion and premise
:param conclusion:
:param premise:
:return: Jacard similarity score
"""
conclusion_tokens = paper_tokenizer(remove_punctuations(conclusion['conclusion_text']).lower())
premise_tokens = paper_tokenizer(remove_punctuations(premise[1]).lower())
intersection = conclusion_tokens.intersection(premise_tokens)
return len(intersection) / (len(premise_tokens) + len(conclusion_tokens) - len(intersection))
def calculate_nltk_sentiment(conclusion, premise):
"""
Calculates the sentiment with the NLTK SentimentIntesityAnalyzer
:param conclusion:
:param premise:
:return:
"""
cleaned_premise_text = remove_punctuations(premise[1].lower())
return SentimentIntensityAnalyzer().polarity_scores(text=cleaned_premise_text)['compound']
def calculate_textblob_sentiment(conclusion, premise):
"""
Calculates the sentiment with TextBlob
:param conclusion:
:param premise:
:return:
"""
cleaned_premise_text = remove_punctuations(premise[1].lower())
return TextBlob(text=cleaned_premise_text).sentiment.polarity
def compute_embedding(embedding, remove_punctuation: bool, file_name: str):
"""
Computes the embedding with given model for all arguments
:param embedding: Model
:param remove_punctuation: Bool to indicate if punctuation should be removed
:param file_name:
"""
arguments = Arguments()
document_embedding = DocumentPoolEmbeddings([embedding])
embedded_arguments = {}
for argument in arguments.ground_truth_arguments:
premises = argument['premises']
conclusion = argument['conclusion']
conclusion_text = conclusion['conclusion_text']
if remove_punctuation:
conclusion_text = remove_punctuations(conclusion_text)
conclusion_sentence = Sentence(conclusion_text)
document_embedding.embed(conclusion_sentence)
embedded_conclusion = conclusion_sentence.get_embedding().detach(
).numpy().tolist()
embedded_premises = {}
argument_uid = None
for premise in premises:
premise_text = premise[1]
if remove_punctuation:
premise_text = remove_punctuations(premise_text)
premise_sentence = Sentence(premise_text)
document_embedding.embed(premise_sentence)
embedded_premise = premise_sentence.get_embedding().detach().numpy(
).tolist()
embedded_premises[premise[2]] = embedded_premise
argument_uid = premise[0]
embedded_arguments[argument_uid] = [
embedded_conclusion, embedded_premises
]
save_embedding(embedded_arguments, file_name)
def calculate_tokenwise_sentiment(conclusion, premise):
"""
Calculates the sentiment tokenwise
:param conclusion:
:param premise:
:return:
"""
words = collections.Counter(re.split(' |\n|\t', remove_punctuations(premise[1].lower())))
sentiment = 0.0
for word in words:
sentiment += words[word] * TextBlob(
text=word).sentiment.polarity # SentimentIntensityAnalyzer().polarity_scores(text=word)['compound']
return sentiment
def calculate_sentiment(self, conclusion, premise):
"""
Calculates the sentiment of the premise
:param conclusion:
:param premise:
:return:
"""
words = collections.Counter(re.split(' |\n|\t', remove_punctuations(premise[1].lower())))
sentiment = 0.0
for word in words:
if self.sentiWordNet_positives.get(word) is not None:
sentiment += words[word] * self.sentiWordNet_positives[word]
if self.sentiWordNet_negatives.get(word) is not None:
sentiment -= words[word] * self.sentiWordNet_negatives[word]
return sentiment