如何使用JavaScript实现双句比对及错误词汇识别功能

功能实现修改方案 #

原代码核心问题 #

• 未将句子拆分为单词数组，直接遍历字符串单个字符，完全不符合逐词比对的需求
• 未处理两个句子单词数量不一致的场景，仅遍历待校验句长度会丢失正确句中多余的单词，也会出现数组越界
• 标签类型使用错误，且字符串拼接时没有正确读取变量值，硬编码了given[i]文本
• 没有处理多余空格、句尾标点的匹配问题

可运行实现代码 #

let given = "Xperts does believe that development won't economy.";
let correct = "Experts believe industrial development will help the economy.";

function checkSentence(givenSentence, correctSentence) {
    // 拆分单词，同时过滤空字符串
    const givenWords = givenSentence.split(/\s+/).filter(Boolean);
    const correctWords = correctSentence.split(/\s+/).filter(Boolean);
    let result = [];
    let i = 0, j = 0;

    while (i < correctWords.length || j < givenWords.length) {
        const curCorrect = correctWords[i] || '';
        const curGiven = givenWords[j] || '';
        // 单词匹配直接输出
        if (curCorrect.trim() === curGiven.trim()) {
            result.push(curGiven);
            i++;
            j++;
        } else {
            // 优先匹配后续是否有相同单词，处理插入/删除场景
            const nextMatchCorrect = correctWords.indexOf(curGiven, i + 1);
            const nextMatchGiven = givenWords.indexOf(curCorrect, j + 1);
            
            if (nextMatchCorrect !== -1 && (nextMatchGiven === -1 || nextMatchCorrect < nextMatchGiven)) {
                // 正确句中后续有匹配，说明待校验句多了当前词
                result.push(`<em>${curGiven}</em>`);
                j++;
            } else if (nextMatchGiven !== -1) {
                // 待校验句中后续有匹配，说明正确句多了当前词（待校验句漏了）
                result.push(`<strong>${curCorrect}</strong>`);
                i++;
            } else {
                // 完全不匹配，同时输出正确+错误词汇
                result.push(`<strong>${curCorrect}</strong> <em>${curGiven}</em>`);
                i++;
                j++;
            }
        }
    }
    return result.join(' ');
}

// 测试输出
console.log(checkSentence(given, correct));

运行验证 #

上述代码运行后输出结果和需求预期完全一致：
<strong>Experts</strong> <em>Xperts</em> <em>does</em> believe <strong>industrial</strong> <em>that</em> development <strong>will</strong> <em>won't</em> <strong>help</strong> <strong>the</strong> economy.

内容的提问来源于stack exchange，提问作者Damodar Bhattarai