Method of Profanity Detection Using Word Embedding and LSTM

Table 1

Separation of the onset, nucleus, and coda.

---

:>>> def run (x)

---

>>>  consonant_ord_list = [ord (char) for char in Korean consonants]

>>>  choseong_list = [char for char in Korean consonants]

>>>  jungseong_list = [char for char in Korean vowels]

>>>  jongseong_list = [char for char in Korean consonants and double consonants]

>>>  result = []

>>>  for char in s:

>>>   if ord (char) = = 32:

>>>    result.append (char)

>>>   elif 48 ≤ ord (char)≤57:

>>>    result.append (char)

>>>   elif consonant_list.count (char) = = 0:

>>>    character_code = ord (char)

>>>    if (55203 < character_code or character_code <44032):

>>>     continue

>>>    code = 44032

>>>    choseong_index = (character_code−code)//21//28

>>>    jungseong_index = (character_code−code−(choseong_index 21 28))//28.

>>>    jongseong_index = character_code−code−(choseong_index 21 28)−(jungseong_index28)

>>>    result.append (choseong_list [choseong_index])

>>>    result.append (jungseong_list [jungseong_index])

>>>    result.append (jongseong_list[jongseong_index])

>>>   else:

>>>    choseong_index = consonant_ord_list.index(ord(char))

>>>    result.append(choseong_list[choseong_index])

>>>    result.append (“−”)

>>>    result.append (“−”)

>>>   return “” .join (result)

---