Markov bot based on google search history.

google_markov.py

import os
import json
import pickle
import random
 
# making a markov bot based on my search history. 21k searches total
# markov code borrowed entirely from:
# http://stackoverflow.com/questions/5306729/how-do-markov-chain-chatbots-work
def parse_searches():
    corpus = open('corpus.txt', 'wt')
    cwd = os.getcwd()
    os.chdir(cwd + "/Searches")
    file_list = os.listdir()
 
    for search_file in file_list:
        with open(search_file) as json_file:
            json_data = json.load(json_file)
            for search in json_data['event']:
                print(search['query']['query_text'], file=corpus)
    print('parsed all searches')
    os.chdir(cwd)
    corpus.close()
 
 
def train():
    print('training markov bot')
    b = open('corpus.txt')
    text = []
    for line in b:
        for word in line.split():
            text.append(word)
    b.close()
    textset = list(set(text))
    follow = {}
    for l in range(len(textset)):
        working = []
        check = textset[l]
        for w in range(len(text) - 1):
            if check == text[w] and text[w][-1] not in '(),.?!':
                working.append(str(text[w + 1]))
        follow[check] = working
    a = open('bot', 'wb')
    pickle.dump(follow, a, 2)
    a.close()
    print('successfully trained')
 
 
def nextword(a, successorlist):
    if a in successorlist:
        return random.choice(successorlist[a])
    else:
        return 'the'
 
 
def bot():
    print('Ask me a question followed by return "quit" to quit')
    a = open('bot', 'rb')
    successorlist = pickle.load(a)
    a.close()
    speech = ''
    while speech != 'quit':
        speech = input('>')
        s = random.choice(speech.split())
        response = ''
        for i in range(random.randint(4, 8)):
            neword = nextword(s,successorlist)
            response += ' ' + neword
            s = neword
        print(response)
 
if __name__ == '__main__':
    parse_searches()
    train()
    bot()