yahoo apiを使って、文章を日本語形態素解析をおこなうスクリプト

README.md

最初に

このスクリプトを動作させるには、２つのライブラリのインストールが必要です。

• beautifulsoup4
• lxml

インストール方法

pip3 install beautifulsoup4 lxml

動作要件

python3 ~

参考サイト

http://qiita.com/IshitaTakeshi/items/dcc8d9271fb14c0a6f58
https://github.com/IshitaTakeshi/NaiveBayes

config.py

# -*- coding: utf-8 -*-
# from https://github.com/IshitaTakeshi/NaiveBayes/blob/master/src/config.py

import os
from configparser import ConfigParser

class Config(object):
    def __init__(self, filename, section):
        if not(os.path.exists(filename)):
            raise ValueError("{} does not exist".format(filename))

        parser = ConfigParser()
        parser.read(filename)

        config = parser.items(section)
        config = dict(config)

        for key, item in config.items():
            config[key] = eval(item)

        #set params as attributes
        self.__dict__ = config

main.py

# -*- coding: utf-8 -*-

# from https://github.com/IshitaTakeshi/NaiveBayes/blob/master/src/splitter.py
# need `pip3 install beautifulsoup4 lxml`

from pprint import pprint
from urllib.parse import urlencode
from urllib.request import urlopen

from bs4 import BeautifulSoup

from config import Config


config = Config('settings.cfg', 'YAHOO')

pageurl = "http://jlp.yahooapis.jp/MAService/V1/parse"

results = "ma"
filter_ = "1|2|3|4|5|9|10"

def split(sentence):
    params = urlencode({'appid': config.appid,
                        'results': results,
                        'filter': filter_,
                        'sentence': sentence})
    params = bytes(params, encoding='utf-8')

    responce = urlopen(pageurl, params)

    # soup = BeautifulSoup(responce.read(), "lxml")
    # return [w.surface.string for w in soup.ma_result.word_list]

    # OR

    # return responce.read().decode("utf-8")


if __name__ == '__main__':
    s = "庭にわ二羽の鶏がいる"
    pprint(split(""))

settings.cfg

[YAHOO]
# obtain from http://developer.yahoo.co.jp/webapi/jlp/ma/v1/parse.html
appid = 'your yahoo app id'