Question

我有一份用户调查文件：

Score    Comment
8        Rapid bureaucratic affairs. Reports for policy...
4        There needs to be communication or feed back f...
7        service is satisfactory
5        Good
5        There is no
10       My main reason for the product is competition ...
9        Because I have not received the results. And m...
5        no reason

我想确定哪些关键字对应较高的分数，哪些关键字对应较低的分数。

我的想法是构造一个单词表（或“单词向量”词典），其中将包含与其相关的分数以及该分数与该句子相关联的次数。

类似以下内容：

Word        Score   Count
Word1:      7       1
            4       2
Word2:      5       1
            9       1
            3       2
            2       1
Word3:      9       3
Word4:      8       1
            9       1
            4       2
...         ...     ...

然后，对于每个单词，平均得分是该单词与之关联的所有得分的平均值。

为此，我的代码如下：

word_vec = {}
# col 1 is the word, col 2 is the score, col 3 is the number of times it occurs

for i in range(len(data)):
    sentence = data['SurveyResponse'][i].split(' ')
    for word in sentence:
        word_vec['word'] = word
        if word in word_vec:
            word_vec[word] = {'Score':data['SCORE'][i], 'NumberOfTimes':(word_vec[word]['NumberOfTimes'] += 1)}
        else:
            word_vec[word] = {'Score':data['SCORE'][i], 'NumberOfTimes':1}

但是此代码给我以下错误：

File "<ipython-input-144-14b3edc8cbd4>", line 9
    word_vec[word] = {'Score':data['SCORE'][i], 'NumberOfTimes':(word_vec[word]['NumberOfTimes'] += 1)}
                                                                                                  ^
SyntaxError: invalid syntax

有人可以告诉我正确的方法吗？

Answer 1

尝试这段代码

word_vec = {}
# col 1 is the word, col 2 is the score, col 3 is the number of times it occurs

for i in range(len(data)):
    sentence = data['SurveyResponse'][i].split(' ')
    for word in sentence:
        word_vec['word'] = word
        if word in word_vec:
            word_vec[word]['Score'] += data['SCORE'][i] # Keep accumulating the total score for each word, would be easier to find the average score later on
            word_vec[word]['NumberOfTimes'] += 1
        else:
            word_vec[word] = {'Score':data['SCORE'][i], 'NumberOfTimes':1}

要增加'NumberOfTimes'的值，您可以像这样word_vec[word]['NumberOfTimes'] += 1

直接增加

Answer 2

您可以使用收集计数器。它可以计算每个单词的出现次数。

这里有个例子：

 from collections import Counter

 c = Counter(["jsdf","ijoiuj","je","oui","je","non","oui","je"])

 print(c)

结果：

Counter({'je': 3, 'oui': 2, 'ijoiuj': 1, 'jsdf': 1, 'non': 1})

您从文档中提取单词并将其放入列表中。最后，该列表将由计数器处理以计算每个单词的出现次数。

如何为每个句子分数计算一个句子中每个单词出现的次数？

2 个答案: