Question

So I am traversing an BinaryTree and the nodes contain strings as well as if the string appeared more than once while reading the file. I am only looking for the top 10 words that appeared most while reading the file, so essentially I am just comparing an int value.

My problem is I am trying to figure out a way to have an effective way to compare and insert a new node if their count value is greater. So say
I have a tree with…

   5
  /  \
 3    10
/       \

1 15

Say the array size is only size 3. I start at 1, since the array is empty till 5 it will look like after hitting 5…

[1],[3],[5]

When I reach 10 it is greater than everything but I would like to keep it sorted so I need to shift 3 to 1, 5 to 3, and 10 to 5. I want to know if there is a more effective way to do this then to shift after every higher number. It reads a text file with over 10k+ words, so I would like it to be as fast as possible.

If a different data structure would be better for this please let me know. I was thinking a queue or linkedlist but I figured array was less wasted space because it’s only size of 10. I am a student so be gentle too :-x.

Answer 1

One way would be to use a couple of data structures. A suggested way, using a Map and a Min-Heap would be as follows:

• As you are reading words from the file, maintain its count in the hash map (word, word_count).
• Maintain a min-heap of size 10. Whenever you update a word in the map, increase its count. Now compare this count with the top element of your min heap. If word_count > value_at_top, then replace the top element and heapify.
• Once you’re done with reading from the file, this heap will contain the top 10 most frequent element in the heap