Basic example of natural language

natural_language_example.py

# http://nltk.org/book
# https://spacy.io/docs/usage/showcase

import nltk
import random

def get_dataset():
    boys = ["Ryan", "Azooz", "Jonard", "Chippo", "Pierre",
            "John", "Michael", "Jordan", "Bobby", "Stephen"]
    girls = ["Cypress", "Ammi", "Kathleen", "Ruby", "Jelly",
            "Jazzel", "Sarah", "Nicole", "Jenna", "Concepcion"]

    dataset = []
    for boy in boys:
        # name, label
        dataset.append((boy,  "Boy"))
    for girl in girls:
        dataset.append((girl,  "Girl"))

    random.shuffle(dataset)
    return dataset

def extract_gender_feature(data):
    data = data.lower()
    feature = {}
    feature["last letter"] = data[-1]
    feature["size"] = len(data)
    feature["vowel count"] = ( data.count("a") +
                               data.count("e") +
                               data.count("i") +
                               data.count("o") +
                               data.count("u"))
    for letter in "abcdefghijklmnopqrstuvwxyz":
        if data.count(letter) == 0:
            continue
        key = "{} count".format(letter)
        feature[key] = data.count(letter)

    return feature

dataset = get_dataset()
features = []
for name, label in dataset:
    feature = (extract_gender_feature(name), label)
    features.append(feature)

classifier = nltk.NaiveBayesClassifier.train(features)

def classify(name):
    print(classifier.classify(extract_gender_feature(name)))

# python -i natural_language_example.py
# >>> classify("Ben")