如何在用户输入中识别名称（来自文本文件）然后打印名称

Question

我的理想目标是让聊天机器人识别出您正在谈论您的一个兄弟姐妹。因此，当您提到您的兄弟或姐妹时（通过名字或关键字：my brother/sister），聊天机器人已经根据文本文件中提供的数据知道他们是谁。我已经弄清楚了关键字部分，但是当我提到他们的名字时（例如：I can't stand James）。聊天机器人不打印我想要的内容，当用户告诉聊天机器人他们兄弟姐妹的名字时，它最终打印了 ("Oh, so your brother's name is") 和 ("I'll make sure to remember that, so what about " + brother_status['name'] + "?")。我该怎么做才能解决这个问题？

我已经试过了，但这似乎也不起作用：

import string

user_input = raw_input("Please enter your sisters name: ").translate(string.maketrans("",""), string.punctuation)    

with open('file.txt') as sibling_database:
if len(user_input.split()) >= 2:
    for line in sibling_database:
        for word in line.split(':'):
            for words in user_input.split():
                if words in word:
                    print("Oh, so your brother's name is " + line.split(':')[1])

这也是我的原始代码（如果您想进行任何其他更改）：

 import string

 brother_status = dict([
    ('name', ''),
    ('nickname', ''),
    ('current age', ''),
    ('relationship', '')])

 brother_keywords = ["Brother", "brother"]
 sister_keywords = ["Sister", "sister"]

def main(): 
while True:
    user_input = raw_input("What type of sibling do you have: ").translate(string.maketrans("",""), string.punctuation)
    for keyword in brother_keywords:
        if keyword in user_input:
            with open('file.txt', 'r') as sibling_database:
                for brother_data in sibling_database:
                    if brother_data.startswith("Brother's Name"):
                        print (brother_data.split(':')[1])   
                        return
                        break

        if user_input.split():
            with open('file.txt') as sibling_database:
                for line in sibling_database:
                    for word in user_input.split():
                        if word in line:
                                print("Oh, so your brother's name is " + line.split(':')[1] * 1)
                                return
                                break


        if user_input.split():
            for keyword in brother_keywords:
                if keyword in user_input:
                    if user_input not in brother_status:
                        with open ('file.txt') as sibling_database:
                            first = sibling_database.read(1)
                            if not first:
                                print ("You never mentioned a brother. What's his name?")
                                user_input = raw_input("What's his name: ")
                                brother_status['name'] = (user_input)
                                with open('file.txt', 'w') as sibling_database:
                                    sibling_database.write("Brother's Name:" + brother_status['name'] * 1 + '\n')
                                    print ("I'll make sure to remember that, so what about " + brother_status['name'] + "?")
                                    continue

  if __name__ == "__main__":
              main()

Answer 1

我已经使用 nltk pos 标签更新了您的代码，以从文本文件中提取名称。试一试：

import string
import nltk

nltk.download('maxent_treebank_pos_tagger')

brother_status = dict([
('name', ''),
('nickname', ''),
('current age', ''),
('relationship', '')])

brother_keywords = ["Brother", "brother"]
sister_keywords = ["Sister", "sister"]

def main(): 
    while True:
    user_input = raw_input("What type of sibling do you have: ").translate(string.maketrans("",""), string.punctuation)
    for keyword in brother_keywords:
        if keyword in user_input:
            with open('file.txt', 'r') as sibling_database:
                data = sibling_database.readline()
                data = nltk.word_tokenize(data)
                tagged_data = nltk.pos_tag(data)
                for name, pos in tagged_data:
                    if pos == "NNP":
                        print(name)
                        return
                        break

        if user_input.split():
            with open('file.txt') as sibling_database:
                for line in sibling_database:
                    for word in user_input.split():
                        if word in line:
                                print("Oh, so your brother's name is " + line.split(':')[1] * 1)
                                return
                                break


        if user_input.split():
            for keyword in brother_keywords:
                if keyword in user_input:
                    if user_input not in brother_status:
                        with open ('file.txt') as sibling_database:
                            first = sibling_database.read(1)
                            if not first:
                                print ("You never mentioned a brother. What's his name?")
                                user_input = raw_input("What's his name: ")
                                brother_status['name'] = (user_input)
                                with open('file.txt', 'w') as sibling_database:
                                    sibling_database.write("Brother's Name:" + brother_status['name'] * 1 + '\n')
                                    print ("I'll make sure to remember that, so what about " + brother_status['name'] + "?")
                                    continue

  if __name__ == "__main__":
      main()

Answer 2

您更大的问题是管理程序的状态。每个循环；您正在测试所有 if，如果它们为真，将一直执行，这不是您想要的。

我建议不要操之过急，例如我认为 if not first: 没有达到您的预期。

帮助组织和管理该状态的一种方法是使用函数。大量使用它们！

然后我建议逐条进行：您需要弄清楚您希望每个 question/answer 在什么条件下出现在您的代码中。如果你问的是一个你不认识的兄弟，那么谈论一个不知名的兄弟的代码可能不应该出现在这个地方。或者应该有条件来保护代码不被执行。

你可能会遇到到处都有条件的情况，当这种情况发生时（而不是之前，或者出于好奇）你应该查看 "state machines"。

关于 python 3 的附注：

Python 2 将在 2020 年过时，不应再使用。系统不会附带它们，人们预计会使用 python 3，对 python 2 的支持将停止。这并不是说您不应该继续使用 python 2 作为学习工具，而是您应该考虑学习 python 3，您会更容易获得更多帮助。还有一些 python 2 没有的很酷的功能，你不会想错过的^^

如何在用户输入中识别名称（来自文本文件）然后打印名称

How to recognize name (from text file) in user input and then print name

python

python-2.x

text-files

chatbot