如何计算对话中每个角色所说的单词数并将计数存储在字典中？

Question

我正在尝试计算字符 "Michael" 和 "Jim" 在以下对话中说出的单词数，并将它们存储在类似于 {"Michael:":15, "Jim:":10} 的字典中。

string = "Michael: All right Jim. Your quarterlies look very good. How are things at the library? Jim: Oh, I told you. I couldn’t close it. So… Michael: So you’ve come to the master for guidance? Is this what you’re saying, grasshopper? Jim: Actually, you called me in here, but yeah. Michael: All right. Well, let me show you how it’s done."

我想创建一个包含字符名称作为键的空字典，将字符串按" " 拆分，然后通过使用键作为参考来计算字符名称之间结果列表元素的数量，然后存储计数词作为值。这是我目前使用的代码：

dict = {"Michael:" : 0,
        "Jim:" : 0}

list = string.split(" ")

indices = [i for i, x in enumerate(list) if x in dict.keys()]
nums = []
for i in range(1,len(indices)):
    nums.append(indices[i] - indices[i-1])
print(nums)

结果是一个打印为 [15, 10, 15, 9] 的列表

我想我需要以下帮助：

1. 如果可能的话，一个更好的方法
2. 当该行是对话的最后一行时，一种计算角色说出的单词数的方法
3. 一种通过自动计算角色说出的单词数来更新字典的方法

最后一点至关重要，因为我试图复制这个过程以获得一集的引语。

提前谢谢你！

Answer 1

遍历单词，不断增加适当的计数。

dialogue_dict = {"Michael:" : 0, "Jim:" : 0}

words = string.split(" ")
current_character = None
for word in words:
    if word in dialogue_dict:
        current_character = word
    elif current_character:
        dialogue_dict[current_character] += 1

顺便说一句，不要使用 list 和 dict 作为变量名，这会用这些名称覆盖内置函数。

Answer 2

• 使用regex to split by character names, keeping the character separators,

• 然后使用 chunks of 2 迭代字符/行对。

  • 使用collections.defaultdict(int)在0处自动添加一个新字符并为当前行添加单词split，

string_ = "Michael: All right Jim. Your quarterlies look very good. How are things at the library? Jim: Oh, I told you. I couldn’t close it. So… Michael: So you’ve come to the master for guidance? Is this what you’re saying, grasshopper? Jim: Actually, you called me in here, but yeah. Michael: All right. Well, let me show you how it’s done."

import re
from collections import defaultdict

#This assumes a character name has no blanks and is followed by a `:`
pat = re.compile("([A-Z][a-z'-]+:)")

#splitting like returns the delimeters (characters) as well
li = [v for v in pat.split(string_) if v]

# split 2 by 2
def chunks(l, n):
    n = max(1, n)
    return (l[i:i+n] for i in range(0, len(l), n))

#use a defaultdict to start new characters at 0
#collections.Counter could also work
counter = defaultdict(int)

pairs = chunks(li,2)
for character, line in pairs:
    counter[character.rstrip(":")] += len(line.split())
 
print(f"{counter=}")

输出：

counter=defaultdict(<class 'int'>, {'Michael': 38, 'Jim': 17})

Answer 3

我们可以使用正则表达式来做到这一点。无需提供演讲者姓名

import re

string = "Michael: All right Jim. Your quarterlies look very good. How are things at the library? Jim: Oh, I told you. I couldn’t close it. So… Michael: So you’ve come to the master for guidance? Is this what you’re saying, grasshopper? Jim: Actually, you called me in here, but yeah. Michael: All right. Well, let me show you how it’s done."
dialog_count = {}

#extract speakers using regex
speakers = re.findall(r'\w+:',string)
#split sentences using regex
sentencs = re.split(r'\w+:',string)
speakers = filter(lambda x: x.strip()!='' ,speakers)
sentencs = filter(lambda x: x.strip()!='' ,sentencs)

#remap each speaker to it's sentence
dialogs = zip(list(speakers),list(sentencs))

#count total words
for speaker,dialog in dialogs:
    dialog = dialog.split(" ")
    dialog = list(filter(lambda x: x.strip()!='',dialog))
    dialog_count[speaker] = dialog_count.get(speaker,0) + len(dialog)
print(dialog_count)

{'Michael:': 38, 'Jim:': 17}