Question

For simplicity, let’s say that I have two sets of words, sorted into alphabetical order. One set starts at “aardvark” and ends at “melon”, and the other starts at “melon” and ends at “zebra”. The word “melon” appears in both sets.

If I were to take an input word, say “banana”, what would be a good (and efficient) way of determining which set of words it should belong to? Note: this isn’t a question about whether the word “banana” already exists in one set, but rather a question about how to determine which set the word should exist in.

If there is an algorithm that someone knows, great. If they can provide some version in Java, even better!

Edit: Should also point out, whilst my example only has 2 sets, I want the algorithm to work with n sets.

Answer 1

For two sets:

If word is your word (e.g. "banana"):

int cmp = word.compareTo("melon");
if (cmp < 0) {
  // it belongs to the first set
} else if (cmp > 0) {
  // it belongs to the second set
} else {
  // the word is "melon"
}

For n sets:

Place the dividing words into an ArrayList<String> (call it dividers) in alphabetical order:

ArrayList<String> dividers = new ArrayList<String>();
//... populate `dividers` ...
Collections.sort(dividers);

Now you can use Collections.binarySearch() to figure out which set the word belongs to:

int pos = Collections.binarySearch(dividers, word);
if (pos >= 0) {
  // the word is the divider between sets `pos` and `pos+1`
} else {
  int num = -(pos + 1);
  // the word belong to set number `num`
}

(Here, the sets are numbered from zero.)