NLP

def extract_column(df, column_name):
  return df[column_name].tolist() # Extract the column 'languages_spacy' from the dataframe
languages_spacy = extract_column(df, 'languages_spacy')
# Print the first 5 rows of the column
print(type(languages_spacy))

def parce_list_of_dictionaries_to_extract_values(languages_spacy):
  return [list(eval(row).values()) for row in languages_spacy]# Parce the values
parced_values = parce_list_of_dictionaries_to_extract_values(languages_spacy)
print(parced_values[0:3])
print(type(parced_values))

def list_to_extract_languages(parced_values):
    return [value[0] for value in parced_values]# Get the languages
languages = list_to_extract_languages(parced_values)
print(languages[0:3])
print(type(languages))



def extract_languages_values_from_df_apply_to_new_column(df, column_name):
  df['languages_spacy_values'] = df[column_name].apply(lambda x: list(eval(x).values()))
  return df

new_df=extract_languages_values_from_df_apply_to_new_column(df, 'languages_spacy')
new_df.head()