使用Javascript来查找字符串中最常见的单词？

Question

我有一大块文字，我想找出最常用的词语（除了少数，如“the”，“a”，“and”等）。

我该如何去寻找这个文本块的最常用的单词？

感谢您的任何想法。

Answer 1

您应该通过的话字符串分割成词，然后循环并且将每一个计数器：

var wordCounts = { }; 
var words = str.split(/\b/); 

for(var i = 0; i < words.length; i++) 
    wordCounts["_" + words[i]] = (wordCounts["_" + words[i]] || 0) + 1; 

的"_" +允许它来处理像constructor词已经是对象的属性。

您可能希望写入words[i].toLowerCase()以便不区分大小写。

Answer 2

从未来开始，再次询问这个问题，但是我对解决方案提前开始并将其标记为已回答。无论如何，这是SLak答案的补充。

function nthMostCommon(string, ammount) { 
    var wordsArray = string.split(/\s/); 
    var wordOccurrences = {} 
    for (var i = 0; i < wordsArray.length; i++) { 
     wordOccurrences['_'+wordsArray[i]] = (wordOccurrences['_'+wordsArray[i]] || 0) + 1; 
    } 
    var result = Object.keys(wordOccurrences).reduce(function(acc, currentKey) { 
     /* you may want to include a binary search here */ 
     for (var i = 0; i < ammount; i++) { 
      if (!acc[i]) { 
       acc[i] = { word: currentKey.slice(1, currentKey.length), occurences: wordOccurrences[currentKey] }; 
       break; 
      } else if (acc[i].occurences < wordOccurrences[currentKey]) { 
       acc.splice(i, 0, { word: currentKey.slice(1, currentKey.length), occurences: wordOccurrences[currentKey] }); 
       if (acc.length > ammount) 
        acc.pop(); 
       break; 
      } 
     } 
     return acc; 
    }, []); 
    return result; 
}