你可以通过下面的步骤来实现：
首先，用regexToUse拆分textToParaphrase字符串，您将得到一个数组，其中textToParaphrase的部分与提供的表达式不匹配。例如：如果，

textToParaphrase = "I am very happy today for you";
 regexToUse = "(very|extremely) (happy|joyful) (today|at this (moment|time|instant in time))";

字符串
输出将为：{"I am ", "for you"}。然后用这些生成的字符串创建一个正则表达式（如"(I am |for you)"）。现在再次用这个生成的表达式拆分textToParaphrase，你将得到给定正则表达式的匹配部分的数组。最后用随机生成的字符串替换每个匹配部分。
代码如下：

public static String paraphraseUsingRegularExpression(String textToParaphrase, String regexToUse){
    String[] unMatchedPortionArray = textToParaphrase.split(regexToUse);
    String regExToFilter = "(";
    for(int i = 0; i< unMatchedPortionArray.length; i++){
        if(i == unMatchedPortionArray.length -1){
            regExToFilter+=unMatchedPortionArray[i];
        } else {
            regExToFilter+=unMatchedPortionArray[i]+"|";
        }
    }
    regExToFilter+=")";

    String[] matchedPortionArray = textToParaphrase.split(regExToFilter);
    Xeger generator = new Xeger(regexToUse);
    for (String matchedSegment : matchedPortionArray){
    String result = generator.generate(); //generates randomly (according to you!)
        textToParaphrase = textToParaphrase.replace(matchedSegment, result);
    }
    return textToParaphrase;
}

型
干杯！干杯！