| def make_token_tag(identifier): | |
| """ | |
| Feature engineering for identifiers, tags each character as a vowel, consonant, number, special character, or other. | |
| Args: | |
| identifier (str): The identifier to tag. | |
| Returns: | |
| str: A string of tag characters the same length as the input string. | |
| """ | |
| vowels = ["a", "e", "i", "o", "u"] | |
| special = ["-", "_", "@"] | |
| numbers = ["1", "2", "3", "4", "5", "6", "7", "8", "9"] | |
| consonants = ["b", "c", "d", "f", "g", "h", "j", "k", "l", "m", "n", "p", "q", "r", "s", "t", "v", "w", "x", "y", "z"] | |
| tags = "" | |
| for c in identifier.lower(): | |
| if c in vowels: | |
| tags += "^" | |
| elif c in special: | |
| tags += "$" | |
| elif c in numbers: | |
| tags += "#" | |
| elif c in consonants: | |
| tags += "+" | |
| else: | |
| tags += "*" | |
| return tags |