找出给定字符序列的下一个字符（逻辑测试样式）（Java）
我最近参加了一个逻辑测验/测试，问题如下：序列的下一个字符是什么：a,c,b,d,c？虽然并不复杂，但我只在规定的时间内完成了其中的一半。
所以我想在下次尝试使用：我构建的脚本或来自互联网的工具。
你有什么想法如何使用java来解决这个问题吗？是否有任何我可以使用或必须从头开始构建的类？我找到了关于 Java Regex Pattern & 的教程Matcher，但我很确定这不是我要找的。
注意：始终是 az chars &通常为 6 组 (+/-1)
I recently took a logic quiz/test with questions like: What is the next character for the sequence: a,c,b,d,c? Although not complicated I only managed to complete like half of them in the given time limit.
So I would like for my next try to use: either a script built by me or a tool from the Internet.
Do you have any ideas how to approach this using java? Are there any classes that I could use or have to build from scratch? I found a tutorial on Java Regex Pattern & Matcher but I'm pretty sure it's not what I am looking for.
Note: It's always a-z chars & usually sets of 6 (+/-1)
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
该序列的合法字母是什么？总是 az 吗？如果是这样，那么预测序列就不那么困难了。您可以将字母映射到 1-26，以获得合理的“猜测器”。
在此示例中：
1, 3, 2, 4, 3...
+2, -1, +2, -1...
您确实需要限定问题以确定解决问题需要多少建模。
What is the legal alphabet for the sequence? Is it always a-z? If so, then predicting the sequence isn't that difficult. You could map the letters to 1-26 for a reasonable 'guesstimator'.
In this example:
1, 3, 2, 4, 3...
+2, -1, +2, -1...
You really need to qualify the question to determine how much modeling is required to solve the problem.
简单问题
在您的情况下，您似乎正在依次选择第 n 个和第 n+2 个字母（以字母表长度为模）以不断生成序列中的下一个字母...该序列可能也会被一些常数稍微错乱......但无论哪种情况，确切的解决方案都应该由人类精确解码并以任何语言实现。
然而，对你的问题的其他评论表明，这个问题暗示了一个完整的、更有趣的问题，人类不容易解决这个问题，而是需要色调学。这个预测问题与生物信息学家和人工智能工程师相关，其中我们想要预测给定前面的单词/字母序列的字符串中的下一个字母或单词（即来自文本流或氨基酸序列）...
完整的问题
这是人工智能中的一个经典问题，需要机器学习。
特定类型的问题将采用
前面的序列作为输入。
并输出：
序列中的单个下一个字符。
github 上有一个氨基酸预测器算法，我们设计该算法是为了使用机器学习来处理这个问题，该算法在 Clojure 中运行（请参阅 jayunit100/Rudolf 项目），如果您对解决这个问题的完整方法感兴趣的话22 个氨基酸的字母表。
The Simple Problem
In your case, it appears you are picking the nth and n+2th letters, in turn (modulo the alphabet length) to continually generate the next letters in a sequence... The sequence might be staggered a little by some constant as well... But in either case, the exact solution should be precisely decoded by a human and implemented in any language.
However, other comments on your question identify that this problem hints at a full blown, much more interesting problem which is not easily solved by a human - but rather which requires hueristics. This prediction problem is relevant to bioinformaticians and artificial intelligence engineers, wherein we want to predict the next letter or word (I.e. From a text stream or Amino acid sequence ) in a string given the preceding word/letter sequence...
The FULL blown Problem
This is a classic problem in artificial intelligence which requires machine learning .
The particular type of problem would take, as input :
the preceding sequence.
And output :
A single, next character in the sequence.
There is an AminoAcid predictor algorithm on github , which we've designed to deal with thus problem using machine learning, that runs in Clojure (see the jayunit100/Rudolf project) , if you are interested in a full blown approach to solving thus problem over a 22 amino acid alphabet.